For Peer Review
Perceptual salience of derivational suffixes in visual word 

recognition

Journal: Scandinavian Journal of Psychology

Manuscript ID SJOP-2019-02-0040.R3

Manuscript Type: Empirical Paper

Keywords: morphological processing, Perceptual suffix salience, Dual route models, 
eye tracking

 

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology



For Peer Review

2

Perceptual salience of derivational suffixes in visual word recognition 

Abstract

This study analyzes the role of derivational suffixes in visual word recognition, tracking 

the eye movements of 31 participants in a sentence-reading task in Spanish. Perceptual 

salience of suffixes was operationalized as the proportion of letters represented by the 

suffixes with respect to the full words, that is, we relate the number of letters 

comprising the suffixes to the number of letters in the words in which they appear. The 

results reveal a significant role in first fixation duration of both word frequency –the 

more frequent the word, the shorter the fixations, and perceptual salience –the more 

salient the suffix, the longer the fixations. Moreover, in gaze duration, our results show 

a main effect of word length –the longer the word, the longer the fixations; word 

frequency; and significant interactions between word frequency and perceptual salience 

of suffixes on the one hand –the effect of word frequency is only significant when 

perceptual salience of suffixes is high, and between word frequency and word length on 

the other hand -the frequency effect decreases as word length increases. Overall results 

are interpreted in the light of the dual route models by which full-form and 

morphological processing interactively cooperate in visual word recognition.

 

Keywords: Dual route models; Eye tracking; Morphological processing; Perceptual 

suffix salience. 

Page 4 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

3

Introduction:

Since Taft and Foster (1975) posited in the 1970s that complex words (those comprising 

at least two morphemes) are morphologically decomposed prior to recognition, much 

progress has been made in understanding how this morphological processing takes 

place. Although the morphological processing of complex words is still not 

unanimously supported in the literature (e.g. Milin, Feldman, Ramscar, Hendrix & 

Baayen, 2017), a large number of studies advocate this possibility (see Amenta & 

Crepaldi for a review, 2012).

The evidence on derivational morphological processing is extensive both with 

regard to the languages analyzed (Dutch, English, Finnish, French, German, Portuguese 

and Spanish among many others), the different variables used to operationalize this 

processing (e.g. base frequency, affix frequency, etc.), and the different methodologies 

implemented. The studies have mainly been conducted using lexical decision or naming 

tasks. These methodologies have several advantages such as the simplicity of the 

technology required, but their results may not necessarily represent what occurs in a 

natural reading (e.g. Hyönä & Pollatsek, 1998), contrary to what it occurs in the more 

ecological sentence reading task. Fortunately, it is possible to study word identification 

in the context of sentence reading by means of eye-tracking devices (e.g. Chumbley & 

Balota, 1984; Bertram, 2011). Therefore word identification can be assessed in a natural 

setting while benefitting from an additional key advantage of the eye tracking. It is that 

this technology allows researchers to obtain data concerning early and late stages of 

processing i.e. it allows to assess the time course of processing (Bertram, 2011).

Eye tracking has been used to gather evidence on the role of morphology in 

word identification with regard to prefixed words (e.g. Beauvillain 1996; Lima, 1987; 

Niswander & Pollatsek, 2006; Pollatsek, Slattery & Juhasz, 2008), suffixed words (e.g. 

Page 5 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

4

Beauvillain 1996; Kuperman, Bertram & Baayen, 2010), and also compound words 

(e.g. Juhasz, 2008; Hyönä & Pollatsek, 1998; Frisson, Niswander-Klement & Pollatsek, 

2008; Inhoff, Starr, Solomon, & Placke 2008; Marelli & Luzzatti, 2012; Pollatsek, 

Hyönä & Bertram, 2000).

For example, in English, Niswander, Pollatsek and Rayner (2000) conducted a 

sentence-reading task in which they manipulated both the word and base frequencies. 

They reasoned that if the results only revealed word-frequency effects, then the 

evidence would suggest that complex words were accessed through their full form, 

without morphological segmentation. Contrastingly, if only base-frequency effects were 

found, then the evidence would corroborate recognition mediated by segmentation and 

morphological processing (e.g. Burani & Thornton, 2003, Sereno & Jongman, 1997, but 

see also Baayen, Wurm & Aycock, 2007). Finally, if both word- and base-frequency 

effects were observed, the case for a dual model would be supported, in which complex 

words are accessed by means of both whole-word processing and morphological 

processing (Schreuder & Baayen, 1995). Their results showed that first fixation duration 

was lower (approximately 17 ms shorter) for high-frequency base words compared to 

low frequency ones. For the same dependent variable, however, the word-frequency 

effect was not statistically significant. Nevertheless, the results for the gaze duration 

variable showed a facilitation effect of word frequency. The authors interpreted the 

overall results under the framework of a dual-route model of processing, stating “the 

data indicate that both the frequency of the root and the frequency of the whole word 

affect the progress of the eyes through the text for derived words” (pag. 411) – see 

Niswander-Klement and Pollatsek (2006) or Pollatsek and Hyönä (2005) for a similar 

theoretical approach.

Page 6 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

5

In French, Beauvillain (1996) also manipulated the variables of word frequency 

and base frequency in two different experiments. However, the experimental task was 

not the classic sentence-reading task, but readers were asked to decide whether or not 

pairs of words were semantically related. They were shown a word on a screen. Later 

another word appeared to its right. Then they were told to make their judgment. The 

results of the first experiment showed that word frequency had no effect on first fixation 

duration although it did affect second fixation duration and gaze duration. In his second 

experiment, Beauvillain manipulated the base frequencies, finding that the high base 

frequency words produced shorter first fixations compared to low base frequency ones. 

The results of these experiments were interpreted as corroborating that full-form 

processing does not constitute part of the first stages of complex word processing while 

morphological segmentation and processing do occur during these early stages. The 

conclusion that morphological decomposition takes place at early processing stages is 

consistent with a large number of studies conducted using both lexical decision and 

naming tasks (e.g. Beyersman, Duñabeitia, Carreiras, & Colheart, 2013; Taft, 2004) as 

well as studies using ERP (e.g. Beyersmann, Iakimova, Ziegler & Colé, 2014) and eye 

tracking (e.g. Amenta, Marelli & Crepaldi, 2015). 

Niswander-Klement and Pollatsek (2006) obtained similar results, but in a 

sentence-reading task in English with prefixed words. These authors also found a 

statistically significant base frequency effect, which was modulated by word length as 

the effect only appeared in long words (mean of 8.5 characters). In contrast, a word- 

frequency effect was only found in the case of short words (mean of 6.5 characters). 

The results presented by Niswander-Klement and Pollatsek (2006) underline the 

possible effect of word length when lexical access is studied using eye movements. In 

this sense, Bertram and Hyönä (2003) developed the “visual acuity hypothesis”. This 

Page 7 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

6

theory holds that the number of characters of a word determines the way a word is 

accessed. Thus, longer, complex words tend to be accessed by means of morphological 

processing, while short words tend to be identified through full-form processing of the 

stimulus (facilitated by word frequency). The key number of characters proposed is 

eight, as it is understood that shorter words can be processed with a single fixation 

while longer words require at least two fixations. 

The length of the complete stimulus is not the only length variable that might be 

important in stimulus processing; suffix length may also be important. Kuperman et al., 

(2010) conducted a sentence-reading study in Dutch with derived words. The authors 

carried out a regression study and found, among other things, that suffix length was an 

explanatory variable. Specifically, they found that the word-frequency effect increased 

when the suffix was short while this effect decreased when the word contained a long 

suffix. The authors associated this effect with suffix perceptual salience. Hence, the 

longer the suffix, the greater is its perceptual salience and the greater is the influence of 

morphologically mediated recognition. In contrast, a lower salience (short) suffix 

promotes a more full-form processing of the stimulus. 

Broadly speaking, all the studies herein reviewed seem to support that the 

properties of morphemes are key in how readers access and recognize complex words. 

In other words, both studies using lexical decision and naming paradigms and those 

conducted with eye tracking methodology corroborate the morphological processing of 

complex words (prefixed, suffixed and compound words). 

However, the literature presents certain limitations that have shaped the 

objectives of this study: 1) the evidence is not extensive and significant gaps and 

questions persist in the literature. For example, as presented above, the results of the 

word frequency in first fixation durations has shown inconsistent results. 2) Few eye-

Page 8 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

7

tracking studies have been conducted on morphological processing, and, of these, most 

are conducted with a factorial design in which the number of factors analyzed is limited 

(see Baayen, 2010, for a defense of regression studies). This is key if we consider the 

dynamic nature of complex word recognition (Marelli & Luzzatti, 2012). 3) The 

majority of eye-tracking studies on morphological processing have been conducted with 

compound words. Far fewer studies have been carried out using derived words and the 

number is even smaller if we consider only those conducted in sentence- reading 

contexts. The importance of the sentence-reading task has already been discussed (see 

Bertram, 2011), but the importance of word type is also worth noting. In Germanic 

languages such as English or Dutch (or Finnish, despite not being a Germanic 

language), word composition is a highly productive procedure for developing new 

words. However, this is not true for Romance languages such as French, Italian and 

Spanish, where derivation is clearly the much more productive procedure (Varela, 

1990). This linguistic difference determines the frequency of one type of word or 

another, as well as the mean length of words, which may have an impact on how they 

are processed. Furthermore, derivation involves the addition of a derivational morpheme 

(of a notably grammatical nature – it tends to change the grammatical category of the 

stem or base, adapting it grammatically to the context of the sentence, etc.) to a lexical 

morpheme (a fully lexical content). However, composition involves the concatenation 

of two lexemes, that is, two lexical morphemes. Evidently, the role of a grammatical 

morpheme can differ from that of a lexical morpheme when processing complex words, 

thus justifying the distinction between composition and derivation. 4) Besides the 

above, the features of each particular language determine other areas of interest for the 

psycholinguistic study of reading. For example, Lázaro and Sainz (2012) or Sainz, 

Sainz and Lázaro (2018) underlined that while in Dutch and English stem family size 

Page 9 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

8

can be in the hundreds – probably due, to a certain extent, to their compositional 

richness – in the corpus of Spanish words in their study, no word had a family size 

above 15. Thus, each language has its own individual characteristics that diverge from 

those of other languages, despite the fact the languages might be related. 

The present study investigates the morphological processing of derived words 

through an analysis of the suffixes and, more specifically, their perceptual salience. 

Perceptual salience is not univocally operationalized in research on morphology and 

lexical access. Salience initially refers to formal aspects, such as the length of suffixes 

measured in graphemes or phonemes, but it can also relate to semantic aspects (Giraudo 

& dal Maso, 2016). Other forms of operationalizing salience exist, such as through 

homonymy,1 (Bertram, Laine, Baayen, Schreuder & Hyönä, 2000) or orthographic 

confusability2 (Laudanna & Burani, 1995). In the present study, however, we refer to 

salience in terms of form, considering salience in relation to the number of characters in 

each suffix, compared to the total number of letters in the word in which the suffix 

appears. Rather than considering a suffix to be more perceptual salient simply because it 

has more graphemes, to determine its perceptual salience, we calculate the proportion of 

the total number of letters in the word represented by the suffix. Thus, a suffix does not 

always exhibit the same perceptual salience, rather, it is more salient in some contexts (-

ista in taxista) than in others (–ista in estructuralista). Although we are aware of 

previous studies reporting the importance of suffix length as regards their association 

with other variables of interest (see Amenta, et al., 2015; Kuperman et al., 2010), we 

consider it appropriate to analyze the perceptual salience of suffixes from this novel 

1 Homonymy refers here to when a suffix has more than one grammatical or semantic function, as in the 
case of –er in “dreamer” [agent] and “higher” [comparative].
2 Confusability refers to the proportion of words included in a letter string that does not constitute a prefix 
or suffix despite coinciding orthographically with one of these forms. For example, “corner” presents the 
pseudosuffix -er. Thus, not all words ending in -er are suffixed words, only a proportion of them. This 
proportion represents the index of confusability.

Page 10 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

9

perspective for a fundamental reason: although the assumption that longer suffixes are 

more perceptually salient than shorter ones is generally accepted, it is also true that in 

many cases this is not so, and, due to the limited number of stimuli used in previous 

studies, the need for precision is paramount. Strictly speaking, perceptual salience 

should include the context, in our case the words. For example, the suffix -ería in 

campechanería (good-naturedness) could be considered less perceptual salient than -al 

in puñal (dagger) despite being twice as long. 

A positive consequence of operationalizing suffix perceptual salience in these 

terms is that a greater variability of values is obtained compared to when perceptual 

salience is defined as the number of letters (in Spanish, derivational suffixes only have 

2, 3 or 4 letters, while our method of calculating perceptual salience in the present study 

yields more than 10 different values, which is significant with regard to the statistical 

analyses conducted). In this sense, the almost totally transparent phonology of Spanish 

means that calculating perceptual salience in phonological terms yields the same results 

as if calculated orthographically, in contrast to the case of other languages. Nonetheless, 

there is an intrinsic difficulty involved in our method of operationalizing salience, 

which is that derivational suffixes in very long words cannot have a high salience level 

while shorter words cannot present low salience. Thus, calculating salience as the main 

factor in our analyses would be contaminated by the fact that words with less salient 

suffixes were longer than words with more salient suffixes. In order to balance the 

levels of high and low salience, we decided to control for the length of the stimuli as 

explained below.

In any event, the aim of this study is not only to examine the role of the 

perceptual salience of suffixes, but also to relate this variable to others with which it 

could potentially interact. For example, from the dual route model approach, it is 

Page 11 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

10

reasonable to posit that suffix perceptual salience has no impact on recognition of high-

frequency words. Furthermore, the effect of perceptual salience of suffixes might be 

expected to be different in short and long words, with its effect being lower in long 

words than in short ones (see Bertram & Hyönä, 2003).

It is worth recalling that all the studies in Spanish conducted using naming and 

lexical decisions find that derivational suffixes play a significant role in lexical 

recognition (e. g. Lázaro, Illera & Sainz, 2015; Lázaro, Acha, de la Rosa, García & 

Sainz, 2016; Lázaro, Acha, Illera, Escalonilla, García, & Sainz, 2018, Suarez-Coalla & 

Cuetos; 2013; Suárez-Coalla; Martínez-García & Cuetos, 2017). Hence, this study 

draws on the notion that, in Spanish, derivational suffixes play a role in the recognition 

of the words of which they form part, and aims to determine to what extent certain 

characteristics of these suffixes, primarily their perceptual salience, contribute to the 

visual recognition of complex words.

Experiment

In this study, a regression analysis was conducted on a sentence-reading task in which 

the readers’ eye movements are monitored using an eye tracker. The aim was to 

determine the effect of suffix perceptual salience on word recognition and how this 

variable interacts with other key variables such as word frequency, suffix frequency and 

word length, among others. Our main prediction is that our novel operationalization of 

the perceptual salience of suffixes is appropriate to assess morphological processing. 

This conclusion would be supported by a significant effect of suffix salience. This effect 

would emerge as a main effect although we also expect an interaction between this 

variable and others such as the word frequency. This interaction would show that the 

larger the salience of suffixes the most prone the lexical system to be sensitive to the 

Page 12 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

11

properties of suffixes. This would imply a reduction of the word frequency effect in the 

case of words with high salient suffixes both in first and gaze durations.

Method

Participants

Thirty-one native Spanish readers participated on a voluntary basis (26 women, 22.1 

mean age). All participants were students of Psychology or Speech and Language 

Therapy at the Complutense University of Madrid. They received course credits for 

their participation. All were native speakers of Spanish with normal or corrected-to-

normal vision.

In order to evaluate the naturalness of the sentences presented, another group of 

17 participants rated sentences in a seven-point Likert scale (from 1, very unnatural, to 

7, very natural). These participants also were students of Psychology and Speech and 

Language Therapy and did not take part in the experiment i.e. they just rated the 

naturalness of the sentences.

Apparatus 

A remote eye-tracking system (ASL, Model 504) was used to record gaze position. The 

system has an accuracy of 0.5º visual angle according the user´s manual. However our 

post hoc calculations have revealed better precision and calibration accuracy. With 

respect to the calibration accuracy the average was 0.5º (SD 0.3º) to the center to each 

calibration point. Diameter of the calibration points was 0.87º. With respect to the 

precision measures the mean distance of the coordinates to the geometric center was 

0.28º and the RMS of the intersample distance was on average 0.17º (SD 0.04º) (see 

Holmqvist, Nyström & Mulvey, 2012). A sampling rate of 50 Hz resulted in the 

Page 13 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

12

coordinates of the gaze being recorded every 20 ms. This rate, being low, is not 

inappropriate to research on eye fixations (see Andersson, Nyström & Holmqvist, 

2010).

The sentences appeared on a screen (166 cm x 122 cm) at a distance of 230 cm 

from the participants. The screen subtended a visual angle of 39º x 29.7º horizontally 

and vertically, respectively. The ocular camera system (94 cm from the participant´s 

eye) was placed such that it caused no interference with the participant’s gaze line to the 

screen. The head position was maintained stable by the use of an anatomical headrest. 

Sessions were carried out individually in a light- and sound-proof Faraday cabin. A 

real-time video scene with a pointer showed the position the participant was looking at. 

Two synchronized computers were used. Data synchronization was possible thanks to 

sixteen extra bits which can be recorded by the program, along with the other eye 

tracker data. Manual responses were recorded using a computer mouse. The luminance 

taken from the eye was kept at a constant level of 57 lux (Digital Lux Meter ISOTECH-

1332A).

Stimuli 

A total of 89 derived words were selected. These stimuli, together with the 31 

participants, will assure more than 2,500 observations which can be considered enough 

to explore the effects of the variables in the light of previous related studies. The 

selected words ended in one of the following 17 suffixes: -ota, -illa, -ura, -eta, -ista, -

eza, -anza, -aje, -al, -iego, -ismo, -ero, -oso, -azo, -izo, -dor, -ez.  To calculate the 

variables, we used Espal, the digital Spanish Lexical Database (Duchon, Perea, 

Sebastián-Gallés, Martí & Carreiras, 2013) and the study by Lázaro, Acha, Illera and 

Sainz (2016), which provided the calculation for suffix frequency in Spanish. To 

Page 14 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

13

calculate perceptual salience of suffixes, the number of letters in the suffix was divided 

by the number of letters in the word and was then multiplied by a hundred. Thus, the 

salience of a suffix would be different according to the number of letters in the word in 

which it appeared (-eza in delicadeza, 3/10 x 100=33,33; -eza in dureza, 3/6 x 100=50). 

As mentioned in the introduction, it was necessary to control for the length of the 

words, avoiding excessively short or long ones so as to ensure a correct distribution of 

suffix salience. As in Burani and Thornton (2003), length was restricted to words 

between 6 and 11 letters.

A total of 89 sentences were generated, into which each of the 89 selected words 

were fitted. All effort was made for the sentences to be equally plausible, minimizing 

any possible plausibility effect, which could have an impact on the morphological 

processing data. As previously mentioned, another group of 17 participants assessed the 

sentences to ensure they presented a high degree of naturalness (see Table 1 for a 

summary of the stimuli and sentences used). The key word was never located at the 

beginning or end of the sentence (Kuperman, Dambacher, Nuthmann & Kliegl, 2010 – 

See Annex1). 

In each trial, one sentence was presented. All sentences occupied one line on the 

screen. The sentences, projected on the center of the screen, subtended a visual angle of 

between 19.08º and 34.25º horizontally and a visual angle of 0.9º vertically. The font 

used in the sentences was Arial 19 and the color was black (0, 0, 0 RGB). Sentences 

appeared on a grey background (160, 160, 160 RGB). Derived words subtended a visual 

angle of between 4.95º and 2º horizontally.

TABLE 1 AROUND HERE

Page 15 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

14

Procedure

The eye tracker was calibrated using a nine-point calibration screen. Data from the right 

eye only were analyzed. In order to ensure the participant was correctly performing the 

reading task, approximately 25% of sentences were followed by a yes-no question 

pertaining to the content of the sentence. The experiment began with a training session 

for the reading task with five sentences and two questions. The sentences used in the 

training session were different from those used in the experimental phase.

Audio instructions, previously recorded, were given before the reading task. The 

main instruction was to read naturally, as normally as possible.

The design was whitin subject; each participant was showed all items. Each trial 

was preceded by a 2000 ms fixation point presented in one character before the 

beginning of the stimulus. The sentences remained on the screen until the participants 

clicked the computer mouse or after a maximum interval of 5000 ms. had elapsed. 

Stimuli were presented randomly. The experiment was programmed using the CVI 

(National Instruments) software. Stimuli were presented randomly.

The experimental sessions lasted 30 minutes at most. 

Dependent variables 

We employed two commonly used variables: 1) the duration of the first fixation, as an 

index of early lexical processing; and, 2) the gaze duration on the derived word, (the 

summed duration of the all fixations on the target word before fixating away from it), as 

it reflects latter stages of processing (Traficante, Marelli & Luzzatti, 2018) and it is an 

index of the processing load at the first encounter with the derived word (Inhoff, 1984; 

Rayner, 1988, Kuperman et al., 2010). This choice was based on the fact that visual 

information is acquired during fixations, when visual processing can be performed with 

Page 16 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

15

high acuity by the fovea and no visual information is acquired during saccades (Rayner, 

1998; Leigh & Zee, 1999). With respect of the first fixation, this is maybe the most used 

measure in eye-tracking (Holmqvist, Nyström, Andersson, Dewhurst, Jarodzka & Van 

de Weijer, 2011). Previous literature has been clear at showing that this measure reflects 

early stages of recognition and that it is sensitive to a number of lexical variables such 

as word frequency for instance (e.g. Niswander, et al., 2000). In the case of the gaze 

duration (a variable also labelled as dwell time) we operativized it as it is the most 

common way (see Holmqvist, et al., 2011). This dependent measured it is hypothesized 

to be sensitive to the cognitive load and has been proved to be very relevant in previous 

studies of morphological processing (e.g. Kuperman et al., (2010).

Predictors 

Table 1 shows a list of the predictors used in the analysis and a summary of their 

characteristics (mean, standard deviation and range). The importance in word 

recognition of the variables introduced in the model has been demonstrated in many 

previous studies. We used word frequency and cumulative root frequency – the summed 

frequency of all derived words sharing the same root (e.g. Giraudo & Grainger, 2000; 

Lázaro; 2012). We also calculated word length, number of neighbors (neighborhood 

density), and family size, a variable whose relevance is supported by specific findings in 

the Spanish language. Suffix frequency, the relevance of which has been reported for 

various languages, including Spanish, was also included in the statistical analysis. The 

number of letters in the word before the key word was also included in the analysis, 

taking into account its possible effect on the key words (Kuperman et al., 2010). 

Finally, the subjective variables of semantic transparency and the naturalness of the 

sentences were also considered.

Page 17 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

16

Data analysis

The analyses were conducted using a set of mixed linear models (Baayen, 2008; 

Baayen, Davidson and Bates, 2008) in R 3.3.2 (2016–10–31) “Sincere Pumpkin Patch” 

(R Core Team, 2016), the lme4 1.1-12 package (Bates, Maechler, Bolker, & Walker, 

2014) and lmerTest 2.0-32 (Kuznetsova, Brockhoff, & Christensen, 2013), with subjects 

and items as random effects. To report the variance explained by the models the MuMin 

package is used (Bartón, 2018). 

Following the analytical procedure proposed by Kuperman, Bertram & Baayen 

(2010), to reduce asymmetry in the distribution of data, logarithmic transformations of 

the dependent variables were conducted and scores outside the range of +3 to -3 

standard deviations from the mean were eliminated. In addition, the continuous 

predictor variables were standardized (subtracting their mean score and dividing by two 

standard deviations, see Kuperman et al., 2010) as a method to facilitate the 

interpretation of the model (Gelman & Hill, 2007). Collinearity is not an issue in our 

initial models as showed in our calculations (Variance Inflation Factor between 1.073 

and 2.288).

An identification by dispersion threshold (IDT) was used as algorithm to identify 

fixations in the study. Several parameters were considered to define a fixation. One of 

them is that fixations shorter than 80 ms. were excluded (Niswander, Pollatsek, & 

Rayner, 2000). Additionally, if standard deviation of the horizontal and vertical eye 

position coordinates was less than 1º visual angle the means of the data sample are used 

as temporary fixation. If standard deviation is greater than this 1º visual angle, then the 

data sample window is moved up one field. Blinks equal or shorter than 200 ms are 

Page 18 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

17

ignored and do not terminate a fixation. Periods longer than 200 ms do cause a fixation 

to close.

The total number of fixations on the target was 4561. We removed fixations 

longer than 1000 ms and more than 3 standard deviations below and above the mean for 

every participant. No fixations were shorter than 80 ms by definition. This yielded a 

total of 4004 valid fixations or fixations that pertained to the first-pass reading. Forty-

six percent of the target words elicited one fixation, 39% elicited two fixations, 12% 

elicited three fixations and only 3% elicited more than three fixations.

Results

The average number of fixations on a stimulus was 1.75 (SD = 0.86). The average 

fixation duration on the target was 235 ms (SD = 101 ms). Considering just the duration 

of the first fixation, the average was 238 ms (SD = 117 ms) and the average gaze 

duration was 408 ms (SD = 273 ms).

All participants responded correctly to at least 90% of the comprehension 

questions (half of the participants responded correctly to all questions, fifteen 

participants answered one question wrongly and only one participant responded 

wrongly to two questions). Therefore, the data on all the participants were used for the 

analysis. 

Because of their low variability and range, sentence naturalness and 

neighborhood density are not included in the model (see Table 1).

 The initial, theoretically-based model for both first fixation and gaze duration 

was the same and included random intercepts for both subjects and items (see Tables 2 

and 3). 

TABLE 2 AND 3 AROUND HERE

Page 19 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

18

The model was simplified using a backward procedure based on the likelihood 

ratio tests. These final models are shown in Tables 4 and 5.

TABLE 4 AROUND HERE

As can be seen in Table 4, with regard to the First Fixation Duration, the 

significant effects involve suffix perceptual salience –the more salient the suffix, the 

longer the fixations; word frequency –the more frequent the word, the shorter the 

fixations; and, with marginal significance, letter length –the longer the word, the longer 

the fixations. 

TABLE 5 AROUND HERE  

As shown in Table 5, with regard to the Gaze Duration, the effect is significant 

for word frequency, with this being modulated by letter length (see Figure 1) –the 

longer the word, the smaller the frequency effect. Suffix salience also interacts 

significantly with word frequency (see Figure 2) –the larger the salience, the greater the 

frequency effect. 

The final models explain, according to r.squaredGLLM command of MuMIn 

library, 7% of variance of the first fixation duration and 16% of the gaze duration. 

FIGURES 1 AND 2 AROUND HERE

Page 20 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

19

Discussion 

This study examines the role of the perceptual salience of suffixes in the processing of 

complex words. This variable is operationalized as the proportion a suffix represents of 

the letter length of the word in which it appears. Furthermore, this variable is related to 

other variables such as word frequency or word length, whose importance in recognition 

of complex words is evidenced in previous studies (e.g. Dahan, Magnuson & 

Tanenhaus, 2001, Juhasz & Rayner, 2006). The study was conducted using eye 

tracking, which permits the analysis of reading processes in a natural task and the 

collection of data that allow the time course of the processing to be recreated. 

The results obtained for first fixation duration reveal a significant effect of word 

frequency, which coincides with the findings of Amenta et al. (2015) o Kuperman et al., 

(2010). This, together with the effect of perceptual salience of suffixes, corroborates the 

dual route model approach in which the system simultaneously computes both lexical 

(frequency) and formal (perceptual salience of suffixes) information at an early stage 

(see Schmidtke, Matsuki & Kuperman, 2017). We support this theory because we 

consider that the effect of the perceptual salience of suffixes implies morpho-

orthographic segmentation in lexical access. To our understanding, a suffix salience 

effect cannot be explained without highlighting that the system is sensitive to the 

morphological composition of the words analyzed, or, in other words, if the reader did 

not process the suffixes as actual suffixes, then no effect of the suffix salience variable 

can be expected. Our interpretation is that the effect of the suffix salience in the first 

fixation durations demonstrates morphological processing of complex words from the 

early processing stages. Thus, and as stated above, an effect of the perceptual salience 

of suffixes together with a word frequency effect in an analysis of first fixation 

durations would support the dual model approach which precisely proposes that the two 

Page 21 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

20

processing routes initiate from early stages (e.g. Duñabeitia, Perea & Carreiras, 2008; 

Frauenfelder & Schreuder, 1992; Hay & Baayen, 2001). 

Continuing the analysis of the suffix perceptual salience effect, we raise the 

question of why the results show that the higher the perceptual salience of suffixes, the 

longer the first fixation durations are. To our view, race models based on the dual route 

model address this issue straightforwardly. These models propose that a direct and a 

compositional route are at play at the same time, in parallel (e.g. Burani & Caramazza, 

1987; Kuperman, Schreuder, Bertram & Baayen, 2009; Schreuder & Baayen, 1995). 

When one becomes predominant depends on the context. For instance Caramazza, 

Laudanna & Romani (1988) proposed that when the word frequency is high the system 

tends to compute the word hollystically. However, in the case of low frequent words, 

the system predominantly processes the stimuli through the compositional route. In the 

same vein, Schreuder and Baayen (1995) considered that the semantic transparency of 

the complex words was the main variable for one or another route to become 

predominant (direct route for opaque words and compositional route for transparent 

words). Traficante et al. (2018) found in this respect relevant results because they 

observed in an eye tracking experiment data consistent with the two routes playing a 

role early in the process. In particular, these authors observed in the first fixation 

durations main effects for base and whole word frequencies. As already said, these 

authors interpreted that these two main variables being significant demonstrated that the 

system computes lexical and compositional routes in parallel and from early stages of 

processing. 

Our assumption goes in the same direction although it points further because it is 

that by increasing the perceptual salience of suffixes readers become more sensitive to 

them and therefore compositional route is favored in comparison to the direct route –or 

Page 22 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

21

in terms of Burani, Marcolini Traficante and Zoccollotti (2018), the longer the 

morphemic constituent is, the more informative this reading unit (see also Kuperman et 

al., 2010, for similar discussion about the interplay between suffix length and word 

frequency). An interesting issue to highlight from the paper by Burani et al., (2018) is 

that they found similar results than us, despite they focused on stem length. Indeed they 

found a facilitative effect by which words with longer stems required less time to be 

read than words with shorter ones. This is in fact what we have observed. Our results 

show that words with high suffix salience (i.e. longer suffixes and shorter stems) 

requires more fixation time than words with low suffix salience (i.e. shorter suffixes and 

longer stems). So, the paper by Burani et al., (2018) and ours are showing two relevant 

aspects of complex word processing. The first one is that by increasing the perceptual 

salience of suffixes readers become more sensitive to them and the second one is that 

the longer the stem –and the shorter the suffix- the faster the processing in the first 

stages. It remains to be explained this second issue, why the effects for stem and suffix 

salience are the opposite?

 To answer this question the recent paper by Sainz, Sainz and Lázaro (2018) can 

be helpful. Sainz et al., (2018) carried out a masked lexical decision task while 

recording EEG confronting the role of stems and derivational suffixes as primes for 

complex words. These authors found a facilitative effect for stem priming as well as an 

inhibitory effect for suffix priming. Sainz et al., (2018), explained the results by a 

probabilistic approach based on the Shannon´s information theory. Basically, these 

authors considered that suffixes are less informative to readers than stems as there are a 

huge number of lexical candidates for a given suffix (as a Romance language derivation 

for Spanish is very productive) while there are few lexical candidates for a given stem 

(as stated above the stem family size for Spanish is very low). Sainz et al., claimed that 

Page 23 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

22

stems are better contributors to lexical recognition than derivational suffixes and that 

this implies that the processing of stems benefit word recognition while presentation of 

suffixes do not benefit lexical recognition, at least in the earliest stages of processing. 

Taken this and all previous research into consideration our interpretation of the effect of 

suffix salience observed in the first fixation times is that in the case of low suffix 

salience, readers are more sensitive to the properties of whole words and stems and 

therefore they process better cues for lexical access in comparison to when complex 

words have higher suffix saliences.   

In short, our interpretation of the results for the first fixation durations suggests 

that: i) the novel operationalization of suffix salience as a proportion between 

morphemes and whole word length is appropriate to research on the role of 

morphological processing in visual word recognition ii) readers become more sensitive 

to the properties of morphemes as their perceptual salience increases iii) the direct –

lexical- processing route and the indirect route, mediated by segmentation and 

morphological processing, operate early and simultaneously and iv) all the above is 

consistent with dual route processing models. 

With regard to gaze duration, the results also show different statistically 

significant effects. One of these refers to the main effect of word length (the longer the 

word is, the more time to be processed). This result cannot come as a surprise given 

previous literature (e.g. Bertram & Hyönä, 2003, Traficante et al. 2018). The effect of 

word length reflects the fact that the system requires more time to extract all the 

information contained in the stimuli when they are large in comparison to when they are 

short. This effect has been typically explained as an index of orthographic processing by 

which compositional route processes letter strings sequentially (Coltheart, Rastle, Perry, 

Langdon & Ziegler, 2001, but see also New, Ferrand, Pallier & Brysbaert, 2006.). 

Page 24 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

23

 Another significant result refers to the interplay between word frequency and 

word length with the frequency effect decreasing as word length increases. This effect 

can be related to the visual acuity hypothesis (Bertram & Hyönä, 2003). To our view, 

this interaction reflects that the system hardly benefits from the word frequency effect 

when the word is large because the visual systems have not enough orthographic 

information to benefit from. In parallel to this result is the one by Bertram & Hyönä 

(2003) in which they observed only a benefit from the word frequency of the first stem 

in long compounds simply because readers “do not have enough letter information on 

the latter part of the word” (p. 625) and therefore readers are not able to benefit from the 

word frequency of the second stem.

More important, in morphological terms, is that despite the lack of significance 

for the main effects of perceptual salience of suffixes and word frequency, the 

relationship between them reaches significance (see Figure 3). The results show that 

when perceptual salience of suffixes is high, the frequency effect is to be expected and 

thus represents a facilitation effect. However, when perceptual salience of suffixes is 

low, this effect is not observed. To understand and explain this result, we find necessary 

to keep in mind four relevant issues. The first one is that it is not possible to abandon 

the idea of the readers being biased towards a morpheme processing when high suffix 

salience occurs. As explained above, our essential assumption concerning the suffix 

salience effect is that it reflects that readers are biased to morpheme processing when 

the perceptual salience of suffixes is large. The second issue is that, as stated in the 

introduction section, the gaze duration does not reflect the same processing stages than 

those supposed for the first fixations. In the case of the gaze duration the results do not 

reflect early stages of processing but the cognitive load and later in-course lexical, 

semantic and morphosyntactic processing. Therefore the interpretation of the results has 

Page 25 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

24

to be related to this integrative processing in the light of the demands of the sentence 

reading task. The third issue is that the explanation by which suffix salience reduces the 

influence of whole word recognition (word frequency) does not hold for these data 

because in that case the interaction would have been in the opposite direction (stronger 

frequency effects for low salient suffixes, see Kuperman et al., 2010). In this respect it 

is important to keep in mind that contrary to previous research we did not consider the 

suffix length, but the proportion of the suffix in the word context as a measure of suffix 

salience. Therefore they are not the same measures –suffix length and suffix proportion- 

and it has important consequences. Our operationalization implies that the same suffix 

length does not suppose the same proportion for every word. Therefore we understand 

that the role of the salience for every suffix can vary according to the word they are 

attached to. For example in the case of the suffix –ez for the words sencillez (simplicity) 

and acidez (acidity) the length is the same but the perceptual salience is higher in the 

latter case. In addition to the measure itself, it is also relevant to consider that our 

stimuli were short in comparison to previous related research conducted in Dutch or 

Finnish (stimuli length ranged from 8 to 14 in Kuperman et al., 2010; and between 12 

and 14 in Hyönä & Pollatsek, 1998; – and between 6 to 11 in our case). The relevance 

of the experimental stimuli length has been shown previously Niswander-Klement and 

Pollatsek (2006). Therefore the results of previous research and ours are framed in a 

different linguistic context and therefore they do not directly challenge ours. The results 

we have observed are better explained considering the next issue: Fourth, the results 

have been obtained in a sentence reading task. In this task the reader is not merely 

required to decide about the lexicality of the stimuli or to read them aloud as in a lexical 

decision or in a naming task, but also to integrate the meaning of the words and relate 

them hierarchically (integrate it grammatically). Specific task demands for readers 

Page 26 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

25

among those tasks are so different that even making use of the same stimuli it is 

possible to find different results. Sentence reading involves a complex demand on the 

reader, who leverages the morphological information to compose the lexical, semantic 

and morpho-syntactic representation of the stimulus. This is of fundamental relevance 

given that derivational suffixes provide formal, lexical, semantic and grammatical 

information that is key to integrating words within the linguistic context formed by a 

sentence that we read and wish to understand. In our view and in the context of the 

previous considerations, the results of the gaze duration are showing precisely the 

contribution of suffixes in performing the sentence reading task. The reader is biased 

towards a morphological processing in the case of high suffix salience but despite not 

being suffixes as informative as stems or whole words in first stages of word 

recognition, they become highly informative later on as they provide morphological 

information to correctly comprehend the sentences. This interpretation is not based on a 

serial model by which a first step (lexical access) is completely done before a second 

step (morphosyntactic processing) is carried out. On the contrary we consider reading as 

a complex task in which different information is able to the reader in a highly 

interactive and dynamic way (Marelli & Luzzatti, 2012; Schmidtke, et al., 2017). 

However, it does not imply that all information entailed in a word or groups of words is 

able at the same time or has the same role for every in course processing. We defend 

that the interaction between word frequency and suffix salience in the gaze duration 

indeed reflects a dynamic processing and shows a cooperative effect why which 

information carried by whole words and suffixes are engaged to facilitate the processing 

and therefore to reduce the cognitive load. Derivational suffixes play a major role in 

relating words of a sentence and as such helpful units, suffixes exert a benefit in 

combination with the frequency of these words. Therefore the benefit from suffixes 

Page 27 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

26

does not necessary preclude or difficult the whole word recognition but it also can 

cooperate thus showing the interaction we have reported. An important issue to 

highlight is then that we defend that lexical and compositional processing are not 

necessarily competitive, on the contrary they can be complementary and cooperate in 

lexical recognition in the context of reading sentences (see Traficante et al. 2018). 

Conclusions

Our study assesses morphological processing in a natural task such as the sentence 

reading task while tracking the eye movements of participants. Our results show 

different interactions between key variables in our two different dependent variables 

thus highlighting the interactive process on word and sentence processing (see 

Schmidtke et al., 2017). In theoretical terms, our overall findings support those models 

that admit that full-form and morphological processing occur early, simultaneously and 

interactively –effects for perceptual salience of suffixes and word frequency in the first 

fixation measure. The results also show, coinciding with Bertram and Hyönä (2003), 

that “the role of morphology in longer words may be determined at least partly by visual 

constraints of the eye” (pag. 619), or, in other words, stimulus length is determinant in 

word recognition. Lastly and importantly, our results support the use of the novel 

variable proposed in this study; the perceptual salience of suffixes, operationalized as 

the proportion between suffixes and whole word length, for studies concerning 

morphological processing. The results obtained support future research on 

morphological processing and perceptual salience of suffixes.

Page 28 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

27

References:

Amenta, S., & Crepaldi, D. (2012). Morphological processing as we know it: an 

analytical review of morphological effects in visual word identification. 

Frontiers in Psychology, 3, 1-12.

Amenta, S., Marelli, M. & Crepaldi, D. (2015). The fruitless effort of growing a

fruitless tree. Journal of Experimental Psychology: Learning, Memory and

 Cognition, 1(5), 1587-1596.

Andersson, R., Nyström. M., & Holmqvist, K. (2010). Sampling frequency and eye

tracking measures: How speed affects durations, latencies, and more. Journal of 

Eye Movement Research, 3, 1-12.

Baayen, H. (2010). A real experiment is a factorial experiment? The mental Lexicon, 5, 

149-157.

Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics

using R. Cambridge University Press.

Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with 

crossed random effects for subjects and items. Journal of Memory and 

Language, 59, 390-412.

Baayen, H., Wurm, L., & Aycock, J. (2007). Lexical dynamics for low-frequency 

complex words. The Mental Lexicon, 2, 419–463.

Bartón, K. (2018). MuMIn: multi-model inference, R Package version 1.9.13.  

Bates, D., Maechler, M., Bolker, B., & Walker, S. (2014). lme4: Linear mixed-effects 

models using Eigen and S4. R package version 1.1-7.

Page 29 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

28

Beauvillain, C. (1996). The integration of morphological and whole-word form

information during Eye fixations on prefixed and suffixed words. Journal of 

Memory and Language 35, 801–820.

Bertram, R., Laine, M, Baayen, H., Schreuder, R., & Hyönna, J. (2000). Affixal 

homonymy triggers full-form storage, even with inflected words, even in a 

morphologically rich language. Cognition 74, 13-25.

Bertram, R., & Hyönna, J. (2003). The length of a complex word modifies the role of 

morphological structure: Evidence from eye movements when reading short and 

long Finnish compounds. Journal of Memory and Language 48, 615–634.

Bertram. R. (2011). Eye movements and morphological processing in reading. The 

Mental Lexicon 6, 83–109. doi 10.1075/ml.6.1.04ber

Beyersmann, E., Iakimova, G., Ziegler, J., & Colé, P. (2014). Semantic processing 

during morphological priming: An ERP study. Brain Research, 1579, 

DOI: 10.1016/j.brainres.2014.07.010

Beyersman, E., Duñabeitia, J.A., Carreiras, M., & Colheart, M. (2013). Early 

morphological decomposition on suffixed words: Masked priming evidence with 

transposed-letter nonword prime. Applied Psycholinguistics, 34, 869-892.

Burani, C., & Caramazza, A. (1987). Representation and processing of derived words. 

Language and Cognitive Processes, 2, 217-227. 

DOI: 10.1080/01690968708406932

Burani, C., & Thornton, A. M. (2003). The interplay of root, suffix and whole-word

frequency in processing derived words. In R. H. Baayen, & R. Schreuder, (Eds.), 

Morphological structure in language processing (pp 157-207). Walter de 

Gruyter.

Page 30 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

29

Burani, C., Marcolini, S., Traficante, D., & Zoccolotti, P. (2018). Reading derived 

words by Italian children with and without dyslexia: The effect of root length. 

Frontiers in Psychology 9, 647. doi: 10.3389/fpsyg.2018.00647

Caramazza, A., Laudanna, A., & Romani, C. (1988) Lexical access and inflectional

morphology. Cognition, 28, 297-332.

Chumbley, J., & Balota, D. (1984). A word's meaning affects the decision in lexical 

decision. Memory & Cognition, 12, 590-606.

Coltheart, M., Rastle, K., Perry, C., Langdon, R., & Ziegler, J. (2001). DRC: A dual 

route cascaded model of visual word recognition and reading aloud. 

Psychological Review, 108, 204-256.

Dahan, D., Magnuson, J., & Tanenhaus, M. (2001). Time course of frequency effects in 

spoken-word recognition: evidence from eye movements. Cognitive Psychology, 

42, 317–367.

Duchon, A., Perea, M., Sebastián-Gallés, N., Martí, A., & Carreiras, M. (2013). EsPal:

          One-stop shopping for Spanish word properties. Behavior Research Methods, 45, 

1246-1258.

Duñabeitia, J.A., Perea, M., & Carreiras, M. (2008). Does darkness lead to happiness? 

Masked suffix priming effects. Language and Cognitive Processes, 23, 1002-

1020.

Frauenfelder, U.H. & Schreuder, R. (1992). Constraining psychological models of 

morphological processing and representation: The role of productivity. In  G. E. 

Booij, and J. van Marle, (eds.), Yearbook of Morphology. (pp. 165-183). Kluwer.

Frisson, S., Niswander-Klement, E., & Pollatsek. A. (2008). The role of semantic

transparency in the processing of English compound words. British Journal of 

Psychology, 99, 87–107.

Page 31 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

http://webpages.ull.es/users/jaduna/Dunabeitia_etal_LCP_2008.pdf
http://webpages.ull.es/users/jaduna/Dunabeitia_etal_LCP_2008.pdf
http://webpages.ull.es/users/jaduna/Dunabeitia_etal_LCP_2008.pdf


For Peer Review

30

Gelman, A., & Hill, J. (2006). Data analysis using regression and 

multilevel/hierarchical models. Cambridge University Press.

Giraudo, H., & dal Maso, S. (2016). The salience of complex words and their parts: 

Which comes first? Frontiers in Psychology. doi: 10.3389/fpsyg.2016.01778.

Giraudo, H., & Grainger, J. (2000). Effects of prime word frequency and cumulative

root frequency in masked morphological priming. Language and Cognitive   

Processes, 15, 421–444.

Hay, J., & Baayen, H. (2001). Parsing and Productivity. In G. Booij and J. van Marie

(eds.), Yearbook of Morphology (pp. 203-235). Kluwer Academic Publishers.

Hyönä, J., & Pollatsek, A. (1998). Reading Finnish compound words: Eye fixations are 

affected by component morphemes. Journal of Experimental Psychology: 

Human Perception and Performance, 24, 1612-1627.

Holmqvist, K., Nyström, M., Andersson, R., Dewhurst, R., Jarodzka, H., & Van de 

Weijer, J. (2011). Eye tracking: A comprehensive guide to methods and 

measures. Oxford University Press.

Holmqvist, K., Nystrom, M., & Fiona Mulvey, F (2012). Eye tracker data quality: What 

it is and how to measure it. Proceedings of the Symposium on eye tracking 

research and application (45-52). DOI: 10.1145/2168556.2168563

Inhoff, A., Matthew, S., Matthew, S., Placke, L. (2008). Eye movements during the 

reading of compound words and the influence of lexeme meaning. Memory & 

Cognition, 36, 675–687.

Inhoff, A. W. (1984). Two stages of word processing during eye fixations in the reading 

of prose. Journal of Memory and Language, 23(5), 612-624.

Juhasz, B. (2008). The processing of compound words in English: Effects of word

Page 32 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

https://books.google.com/books?hl=es&lr=&id=c9xLKzZWoZ4C&oi=fnd&pg=PR17&ots=bbP9OXMmmf&sig=84gwvFQ8GH-3rZnqyUf8rCdziIE
https://books.google.com/books?hl=es&lr=&id=c9xLKzZWoZ4C&oi=fnd&pg=PR17&ots=bbP9OXMmmf&sig=84gwvFQ8GH-3rZnqyUf8rCdziIE


For Peer Review

31

length on eye movements during reading. Language and Cognitive Processes, 

23, 1057-1088. 

Juhasz, B., & Rayner, K. (2006). The role of age of acquisition and word frequency in

reading: Evidence from eye fixation durations. Visual Cognition, 13, 846-863.

Kuperman, V., Schreuder, R., Bertram, R., & Baayen., H. (2009). Reading 

polymorphemic Dutch compounds: toward a multiple route model of lexical 

processing. Journal of Experimental Psychology: Human Perception and 

Performance, 35, 876–895.

Kuperman, V., Bertram, R., & Baayen, H. (2010). Processing trade-offs in reading 

Dutch derived words. Journal of Memory and Language, 62, 83-97.

Kuperman, V., Dambacher, M., Nuthmann, A., & Kliegl, R. (2010). The effect of word

position on eye-movements in sentence and paragraph reading. Quarterly 

Journal of Experimental Psychology. 63, 1838 –1857.

Kuznetsova, A., Brockhoff, P. B., & Christensen, R. H. B. lmerTest: Tests in linear

mixed effects models; 2013. R package version 2.0–20.

Laudanna, A. & Burani, C. (1995).   Distributional properties of derivational affixes: 

Implications for processing. In L.B. Feldman (Ed.), Morphological Aspects of 

Language Processing: Cross-Linguistic Perspectives (pp. 345-364). Hillsdale: 

Lawrence Erlbaum Associates.

Lázaro, M. (2012). A study of base frequency in Spanish skilled and reading-disabled 

children: all children benefit from morphological processing in defining complex 

pseudowords. Dyslexia, 18, 130-138.  DOI: 10.1002/dys.1436.

Lázaro, M., Acha, J., de la Rosa, S., García, S., & Sainz, J. (2016). Exploring the 

derivative suffix frequency in Spanish speaking children. Reading and Writing. 

DOI :10.1007/s11145-016-9668-2.

Page 33 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

http://journals.sagepub.com/author/Kuperman%2C+Victor
http://journals.sagepub.com/author/Dambacher%2C+Michael
http://journals.sagepub.com/author/Nuthmann%2C+Antje
http://journals.sagepub.com/author/Kliegl%2C+Reinhold
http://pubget.com/paper/22337120
http://pubget.com/paper/22337120
http://pubget.com/paper/22337120


For Peer Review

32

Lázaro, M., Acha, J., Illera, V., & Sainz, J. (2016). Written type and token frequency

measures of fifty Spanish derivational morphemes. Spanish Journal of 

Psychology. Doi:10.1017/sjp.2016.75

Lázaro, M., Illera, V., Acha, J., Escalonilla, A., García, S., & Sainz, J. (2018).

Morphological effects in word identification: Tracking the developmental 

trajectory of derivational suffixes in Spanish. Reading and Writing.

Lázaro, M., Illera, V., & Sainz, J. (2015). The suffix priming effect: further evidence for

an early morpho-orthographic process independent of its semantics status. 

Quarterly Journal of Experimental Psychology. 

http://dx.doi.org/10.1080/17470218.2015.1031146.

Lázaro, M., & Sainz, J. (2012). The effect of Family Size on Spanish simple and 

complex words. Journal of Psycholinguistic Research, 41, 181-193.

Leigh, R. J., & Zee, D. S. (1999). The neurology of eye movements. New York: Oxford 

University Press. 

Lima, S. (1987). Morphological analysis in reading. Journal of Memory and Language 

26, 84-99.

Marelli, M., & Luzzati, C. (2012). Frequency effects in the processing of Italian 

nominal compounds: Modulation of headedness and semantic transparency. 

Journal of Memory and Language, 66, 644-664. 

Milin, P., Feldman, L., Ramscar, M., Hendrix, P., & Baayen, H. (2017). Discrimination 

in lexical decision. Plos One. doi:10.1371/ journal.pone.0171935.

New, B., Ferrand, L., Pallier, C., & Brysbaert, M. (2006). Reexamining the word length 

effect in visual word recognition: New evidence from the English Lexicon 

Project. Psychonomic Bulletin & Review, 13, 45-52.

Niswander-Klement, E., & Pollatsek, A. (2006). The effects of root frequency, word

Page 34 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

33

frequency, and length on the processing of prefixed English words during 

reading. Memory & Cognition 34, 685-670.

Niswander, E., Pollatsek, A., & Rayner, K. (2000). The processing of derived and

inflected suffixed words during reading. Language and Cognitive Processes, 15, 

389– 420.

Pollatsek, A., & Hyönä, J. (2005). The role of semantic transparency in the processing

of Finnish compound words. Language & Cognitive Processes, 20, 261-290.

Pollatsek, A., Hyönä, J., & Bertram, R. (2000). The role of morphological constituents 

in reading Finnish compound words. Journal of Experimental Psychology: 

Human Perception and Performance, 26, 820-833 DOI: t0.1037//0096-

1523.26.2.820 2000

Pollatsek, A., Slattery, T.J., Juhasz, B. (2008). The processing of novel and lexicalized

prefixed words in reading. Language and Cognitive Processes, 23, 1133-1158.

Rayner, K. (1998). Eye movements in reading and information processing: 20 years of 

research. Psychological Bulletin,124, 372-422.

Sainz, L., Sainz, J., & Lázaro, M. (2018). Oscillatory brain activity in morphological 

parsing of complex words: Information gain from stems and suffixes. Journal of 

Neurology and Neuroscience, 9,  DOI: 10.21767/2171-6625.1000271.

Schmidtke, D., Matsuki, K., & Kuperman, V. (2017). Surviving blind decomposition: a 

distributional analysis of the time-course of complex word recognition. Journal of 

Experimental Psychology Learning Memory and Cognition 43, 1793-1820. 

DOI: 10.1037/xlm0000411.

Schreuder, R., & Baayen, R. H. (1995). Modelling morphological processing. In L. B. 

Feldman (Ed.), Morphological aspects of language processing (pp. 131-154). 

Hillsdale, New Jersey: Erlbaum.

Page 35 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

https://www.researchgate.net/publication/328743209_Oscillatory_Brain_Activity_in_Morphological_Parsing_of_Complex_Words_Information_Gain_from_Stems_and_Suffixes?_sg=eRjzZ2IxvA0UWaI2nh4eCPbjO5pYhVkYkZX7NFXxQrRvpDwRdQULMOn6BRsHFSBgp7UG69Nu8rtiJfeFKCjd-FKVq9Sz-PtMNvGFjyKx.uiDINuib6wguQ3wZsHJ8gVP2oVhBEKXuj6BLlfshss-JYjqB_CQSakSwiNYBaMPZThprKBdVXrNKVwzvvMxwfQ
https://www.researchgate.net/publication/328743209_Oscillatory_Brain_Activity_in_Morphological_Parsing_of_Complex_Words_Information_Gain_from_Stems_and_Suffixes?_sg=eRjzZ2IxvA0UWaI2nh4eCPbjO5pYhVkYkZX7NFXxQrRvpDwRdQULMOn6BRsHFSBgp7UG69Nu8rtiJfeFKCjd-FKVq9Sz-PtMNvGFjyKx.uiDINuib6wguQ3wZsHJ8gVP2oVhBEKXuj6BLlfshss-JYjqB_CQSakSwiNYBaMPZThprKBdVXrNKVwzvvMxwfQ


For Peer Review

34

Sereno, J. & Jongman. A. (1997). Processing of English inflectional morphology. 

Memory and Cognition, 25, 425-437.

Suárez-Coalla, P., & Cuetos, F. (2013). The role of morphology in reading in Spanish-

speaking children with dyslexia. The Spanish Journal of Psychology, 16, 

doi:10.1017/sjp.2013.58.

Suárez-Coalla, P., Martínez-García, C., & Cuetos, F. (2017). Morpheme-based reading 

and writing in Spanish children with dyslexia. Frontiers in Psychology. 

doi.org/10.3389/fpsyg.2017.01952.

Taft, M. (2004). Morphological decomposition and the reverse base frequency effect. 

Quarterly Journal of Experimental Psychology, 57, 745-765.

Taft, M., & Forster, K. (1975). Lexical storage and retrieval of polymorphemic and

polysyllabic words. Journal of Verbal Learning and Verbal Behavior, 15(6),

607-620.

Traficante, D., Marelli, M., & Luzzatti, C. (2018). Effects of reading proficiency and of 

base and whole-word frequency on reading noun- and verb-derived words: An 

eye-tracking study in Italian primary school children. Frontiers in Psychology. 

doi: 10.3389/fpsyg.2018.02335

Varela, S. (1990). Fundamentos de morfología. Madrid: Síntesis.

Page 36 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

https://www.cambridge.org/core/journals/spanish-journal-of-psychology/article/role-of-morphology-in-reading-in-spanishspeaking-children-with-dyslexia/861B5DB64B12B5AB830ADC14C4D2B6C0
https://www.cambridge.org/core/journals/spanish-journal-of-psychology/article/role-of-morphology-in-reading-in-spanishspeaking-children-with-dyslexia/861B5DB64B12B5AB830ADC14C4D2B6C0
https://doi.org/10.3389/fpsyg.2017.01952


For Peer Review

35

Table 1. Mean, standard deviation and range of experimental stimuli 

Mean SD Range
Word Freq. (per million) 6.3 9.7 0.1 - 54.5
Cumulative Root Freq.  (per million) 67.2 86.7 1.9 - 259
Family Size 5.4 2.7 1 - 12
Sem. Transparency (1 -7) 5.1 2.7 3.08 - 6.4
N. Density .9 1.1 0 - 4
Letter Length 8.7 1.4 6-11
Salience 37.1 8.2 16.7 - 66.7
Suffix Freq. (per million) 1863.8 2664.2 151 - 11,357
Sentence Naturalness (1 -7) 5.4 .4 4.29 - 6.23
Sentence length (words) 11.08 1.65 8 - 17
Letters previous word 3.4 2.01 2 - 11

Page 37 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

36

Table 2. Results of the initial model for the First Fixation Duration

                                                                         Estimate        Std. Error       df              t value       Pr(>|t|)    
(Intercept)                                                        -6.866e-01   9.335e-03   3.640e+01   -73.554   <2e-16 ***
Sem. Transparency                                         -1.402e-02    9.929e-03   9.383e+02    -1.412    0.1582
Word Frequency                                             -3.379e-02   1.608e-02   8.030e+01    -2.101     0.0388 *
Letter Length                                                   1.927e-02    1.400e-02   7.380e+01    1.377     0.1728 
Suffix Salience                                                2.758e-02   1.373e-02   7.160e+01      2.009     0.0483 *
Suffix Frequency                                             1.605e-02    1.434e-02   8.040e+01     1.120    0.2662 
Cumulative Root Freq.                                    -6.402e-03   1.437e-02   7.340e+01    -0.445    0.6573 
Letter Length Previous Word                           7.795e-03    9.214e-03   7.410e+01    0.846    0.4002
Word Freq:Sem.Transparency                         1.484e-02     1.816e-02   1.973e+03    0.817     0.4139
Word Freq:Suffix Freq.                                    2.850e-02    6.123e-02   8.080e+01     0.465     0.6428  
Word Freq:Letter Length                                 1.025e-02     3.425e-02   7.270e+01    0.299     0.7656
Word Freq:Suffix salience                               -3.236e-03   4.970e-02   7.140e+01    -0.065    0.9483
Word Freq: Cumulative Root Freq                  1.054e-02    3.219e-02   7.450e+01    0.327     0.7443 
Word Freq:Suffix Salience:Suffix freq            3.097e-02    7.675e-02   7.960e+01    0.403     0.6877   
Word Freq:Letter Length:Suffix salience       -1.345e-01    6.883e-02   7.470e+01   -1.954    0.0544 .

Page 38 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

37

Table 3. Results of the initial model for Gaze Duration 

                                                                              Estimate     Std. Error         df          t value      Pr(>|t|)    
(Intercept)                                                           -4.868e-01   1.879e-02    3.610e+01  -25.910  < 2e-16 ***
Sem. Transparency                                              -8.305e-03   1.348e-02   1.418e+03  -0.616  0.53797
Word Frequency                                                   -2.781e-02   2.292e-02   8.360e+01  -1.213  0.22846 
Letter Length                                                        6.326e-02    2.056e-02   7.970e+01   3.076  0.00287 **
Suffix Salience                                                     1.976e-02    2.042e-02   7.820e+01   0.968  0.33621
Suffix Frequency                                                  4.213e-03    2.153e-02   8.170e+01   0.196  0.84532 
Cumulative Root Freq.                                         1.386e-02    2.209e-02   7.790e+01   0.627  0.53223
Letter Length Previous Word                               -6.644e-03    1.418e-02   7.800e+01  -0.469  0.64071
Transparency:word frequency                              1.158e-02    2.368e-02    2.148e+03   0.489  0.62490 
Word Frequency:Suffix frequency                       -4.984e-03    9.510e-02   7.850e+01  -0.052   0.9583    
Word Frequency:Letter length                             -8.245e-02   5.289e-02   7.700e+01  -1.559  0.12311 
Word Frequency:Suffix salience                          -1.813e-01    7.289e-02   7.730e+01  -2.487  0.01503 *
Word Frequency: Cumulative Root Freq             2.141e-02    4.958e-02   7.810e+01  -0.432  0.66709
Word Frequency:Suffix salience:Suffix freq        5.151e-02    7.242e-02   7.770e+01   0.711  0.47902
Word Frequency:Letter length:Suffix Salience    -9.706e-02  1.025e-01   7.930e+01  -0.947  0.34648

Page 39 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

38

Table 4. Results of the final model for the First Fixation Duration 

                                                                           Estimate         Std. Error         df            t value       Pr(>|t|)    
(Intercept)                                                           -0.684171   0.008947    31.100000   -76.473   < 2e-16 ***
Sem. Transparency                                             -0.014693   0.009590    788.800000  -1.532     0.12588    
Word Frequency                                                -0.034367   0.012077     83.600000    -2.846     0.00557 **
Letter Length                                                      0.021960   0.011940      82.700000    1.839      0.06947 .  
Suffix Salience                                                    0.027605   0.011399      80.300000    2.422      0.01770 *
Suffix Frequency                                                 0.016128   0.011111      85.500000    1.452      0.15029
Word Freq:SuffixSalience:Letter Length           0.105709   0.055600      83.300000    -1.901     0.06073.  

Page 40 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

39

Table 5. Results of the final model for Gaze Duration 

                                                               Estimate     Std. Error      df          t value     Pr(>|t|)    
(Intercept)                                              -0.48839    0.01836    33.46000   -26.595   < 2e-16 ***
Word Frequency                                   -0.02051    0.01465    83.84000   -1.400     0.165127
Letter Length                                          0.06228    0.01855    85.77000    3.357      0.001177 **
Suffix Salience                                       0.02437    0.01651     82.99000    1.475     0.143873    
Word Freq:Letter Length                       -0.09214    0.04030    85.46000   -2.286     0.024707 *  
Word Freq:Suffix Salience                    -0.18396    0.05084    83.13000   -3.618     0.000508 ***

Page 41 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

40

Figure 1. Interaction between word frequency and letter length. 

Page 42 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

41

Figure 2. Interaction between frequency and suffix salience. Salience is divided into 

four groups to facilitate understanding of the figure. Very high suffix salience >55%, 

high suffix salience >45%, medium suffix salience >35%, low suffix salience >25%,    

Page 43 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

42

Annex 1. Sentences and key words of the experiment

Hermosura    La niña es una completa hermosura vestida con ese trajecito
Adorable     The girl looked absolutely adorable wearing that little dress.
Guionista     El director es bueno pero el guionista es el mejor del equipo
Scriptwriter     The director is good but the scriptwriter is the best person on the team.
Delicadeza   La tigresa cogió a su cría con gran delicadeza por el cuello
Tenderness   The tigress picked up her cub by the neck with the utmost tenderness.
Cabezota       Ese muchacho siempre ha sido muy cabezota para temas de dinero
Obstinate     That boy has always been obstinate in money matters.
Anchura         El carril no tiene la suficiente anchura para que circule ese camión
Width        The width of the lane is insufficient for the lorry.
Belleza         Las montañas nevadas de Canadá son de una belleza extraordinaria 
Beautiful         The snowy mountains in Canada are extraordinarily beautiful. 
Patriota         El soldado siempre ha sido muy patriota aun en los peores momentos
Patriotic         The soldier has always been really patriotic even in the worst of times.
Boquilla         El trompetista tuvo que comprar una boquilla justo antes del concierto
Mouthpiece    The trumpet player had to buy a mouthpiece just before the concert.
Puñal   La escopeta y el puñal han sido siempre armas clásicas entre los bandidos 
Dagger   The shotgun and dagger are the classic weapons of bandits. 
Almohadilla   Le prestaron una almohadilla en el avión y pudo dormir sin problemas
Pillow   He was given a pillow and had a good sleep on the plane.
Alfombrilla    Al salir de la ducha usa la alfombrilla para evitar caerte
Bathmat    When you get out the shower, use the bathmat to avoid any accidents.
Narizota        Sus amigos se reían de su narizota aunque a él le daba igual 
Honker        His Friends laughed at his “honker” of a nose but he didn’t care. 
Espesura        En la selva la espesura de los vegetales es extraordinaria
Density        The density of the vegetation in the rainforest is remarkable.
Avioneta        Se compró una avioneta con el dinero que le pagó el seguro
Aircraft        He bought a small aircraft with the money he got from the insurance.
Extrañeza    El abogado mostró su extrañeza por no haber sido informado formalmente
Astonishment    The lawyer’s astonishment at the lack of prior information was clear.
Mascarilla     Es bueno echarse mascarilla en el pelo para acondicionarlo
Conditioner     It’s a good idea to use conditioner on your hair after shampooing.
Camiseta        Una de las camisetas más bonitas que he visto no vale ni diez euros
T-shirt       It’s one of the nicest t-shirts I’ve ever seen and it only costs ten euros.
Telefonista  Me atendió una telefonista que me recordó a mi hermana
Telephonist  The telephonist who spoke to me reminded me of my sister.
Aspereza      Al modisto no le gustaba la excesiva aspereza del tejido
Roughness      The dressmaker was not happy with the roughness of the cloth.
Gobernanza  El secreto de una buena gobernanza reside en no perder el horizonte
Governance  The secret to good governance is keeping your eye on the horizon.
Andanza     El policía acabó con la andanza del ladrón de forma un tanto brusca  
Adventure     The police abruptly ended the thief’s adventures in crime.  

Page 44 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

43

Colchoneta   Hace los abdominales en una colchoneta que da pena verla 
Matting   He does sit-ups on a worn out piece of matting. 
Capitalista    Hubo un debate sobre los beneficios del sistema capitalista y sus defectos
Capitalist    They held a debate on the pros and cons of the capitalist system
Estructural  El problema de la empresa es estructural y no tiene fácil solución
Structural  The company suffers from a structural problem that cannot be easily solved.
Plomizo     El cielo tiene un color plomizo que contrasta con el verde del campo
Greyish     The greyish color of the sky is in stark contrast to the green countryside.
Estupidez   La mayor estupidez la cometí al aceptar el trabajo
Stupidity   Accepting the job was an act of great stupidity.
Pegajoso     Se trata de un líquido pegajoso bastante tóxico
Sticky    It is a highly toxic sticky liquid.
Peligroso     Es bastante peligroso conducir el coche en ese estado
Dangerous     It is really dangerous to drive in that condition.
Resbaladizo   El suelo queda muy resbaladizo con el rocío de la mañana
Slippery   The morning dew has made the ground really slippery.
Gallinero    Los aficionados suben al gallinero pero todos preferirían subir al palco
Gallery    The fans always sit up in the gallery but would prefer a box seat.
Cuidadoso   Has de ser más cuidadoso con lo que dices a tu madre
Careful   You have to be careful about what you say to your mother.
Embajador   En mi viaje a Japón visité al embajador español
Ambassador   I visited the Spanish ambassador on my trip to Japan.
Patinaje     Se rompió la pierna practicando patinaje sobre hielo
Skating     She broke her leg while doing ice-skating.
Orgulloso    El padre del campeón estaba orgulloso de su hijo
Pride     The father was full of pride for his son.
Mujeriego   Su marido siempre ha sido muy mujeriego aunque no lo reconozca
Womanizer   She refuses to accept it but her husband has always been a womanizer. 
Futbolista    Como estudiante era un desastre pero como futbolista es un genio
Footballer    He was a disastrous student but a great footballer
Coleccionista  Su padre es un coleccionista reconocido en toda Europa 
Collector  Her father is an art collector who is famous across Europe. 
Carruaje     Hace siglos que la reina no usa el carruaje para sus desplazamientos 
Carriage     The queen has not used a carriage as a means of transport for many years. 
Frutero    En mi barrio conozco a un frutero que tiene un Ferrari rojo
Greengrocer    In my neighborhood, I know a greengrocer who drives a Ferrari.
Palabrota      El padre regañó al niño por decir una palabrota muy desagradable
Swearword      The father told the child off for using such a horrible swearword.
Fianza          Si no abonas la fianza no puedes salir de la cárcel
Bail          If you don’t pay the bail, you won’t get out of prison.
Alcoholismo  El problema del alcoholismo ha alcanzado cotas insospechadas
Alcoholism  The problem of alcoholism is worse than ever.
Tramposo   Me enfadé con mi primo porque es muy tramposo jugando al ajedrez
Deceitful   I got angry with my cousin because he is deceitful when he plays chess

Page 45 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

44

Simulador    El piloto entrena con un simulador de vuelo muy avanzado
Simulator    The pilot trains using a state-of-the-art simulator.
Montaje     El acusado dice que todo es un montaje orquestado por su socio
Setup     The accused claims it was all part of a setup implemented by his partner.
Prisionero   La cárcel no está preparada para prisioneros como ese traficante
Prisoner   The jail is not ready for prisoners like that drug trafficker.
Enfermizo    Es un problema su necesidad enfermiza de tener el control de todo
Unhealthy    He has an unhealthy need to be in control all the time.
Ligadura         Como método anticonceptivo la ligadura de trompas es muy efectivo
Ligation         Tubal ligation is a highly effective method of contraception.
Labranza   Las labores propias de  labranza son todas muy duras
Farming   Farming is an extremely hard profession.
Administrador   La página web tiene un administrador que es un desastre
Administrator   The administrator of the website is useless.
Crianza       Confío plenamente en Julieta para la crianza de mi hijo 
Education       I have entrusted Julietta with the education of my child. 
Brutal  La inversión que necesitamos es brutal para igualarnos a Estados Unidos 
Staggering  The investment needed to put us at the level of the USA is staggering. 
Alarmismo   No hay que caer en el alarmismo cuando se informa de cosas así
Alarmist   When we read about things like this, we must avoid alarmist reactions
Cucharilla      Lo mejor es utilizar una cucharilla para echar los condimentos
Spoonful      It is best to use a little spoon to serve condiments.
Bravura          El búfalo es un animal de gran bravura cuando  vive aislado del grupo
Ferocity          The buffalo is an animal of great ferocity when isolated from the group.
Chispazo     Al cambiar la batería le ha dado un chispazo que casi tiene que ir al hospital
Shock   When he changed the battery he got such an electric shock that he almost had to go to hospital.
Acidez    La cerveza que bebes tiene un nivel de acidez que excede el recomendado
Acidity   The beer you’re drinking has higher than recommended acidity.
Pobreza        En los países en desarrollo la pobreza está desapareciendo lentamente
Poverty        Poverty is slowly disappearing in the developing countries.
Adivinanza    A mi hijo le encantan las adivinanzas aunque nunca las adivine
Riddle      My son loves riddles although he never gets them right.
Diseñador    La pasarela contó con la labor de un diseñador y tres modelos
Designer    The fashion parade featured a designer and three models.
Veraniego   El periodo veraniego es el que más disfruto del año
Summer   Summer is my favorite time of the year.
Golpazo      Se cayó al suelo y se dio un golpazo tremendo en la boca
Bang      He fell to the ground and got a terrible bang on his mouth.
Asustadizo  El perro que he adoptado es muy asustadizo y no se deja acariciar
Nervous     The dog he adopted is very nervous and doesn’t like being stroked.
Jardinero   El oficio de jardinero está infravalorado en nuestro país
Gardener   Being a gardener is an underappreciated profession in our country.
Camioneta    Tuvo un accidente con la camioneta que casi le deja cojo
Van    He had an accident with his van and almost lost a leg.

Page 46 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

45

Floral  Se realizó una ofrenda floral en el centenario de su nacimiento 
Floral  The centenary of his birth was marked by a floral offering. 
Documental  Se echa la siesta viendo el documental de la televisión
Documentary  He takes a nap when the documentary starts on TV.
Clasismo   En el siglo diecinueve existía un fuerte clasismo en las esferas más aburguesadas
Classism  In the nineteenth century, there was much classism amongst the bourgeoisie.
Pensador   El premio se lo han dado a un pensador polaco muy famoso
Thinker   The prize was awarded to a famous Polish thinker.
Brillantez   Su currículum es de una brillantez francamente asombrosa
Excellence   The excellence of her CV is frankly breathtaking.
Derrotismo  Con su derrotismo lo que consigue es desanimar al equipo 
Defeatism  Your defeatism only makes the team lose heart.
Nerviosismo  Lo mejor es superar el nerviosismo inicial y responder adecuadamente 
Nervousness  It’s best to overcome any initial nervousness and to reply as well as possible. 
Palaciego    Se tramó un ardid palaciego para expulsar al conde de Marivalle
Courtier     The courtier hatched a plot to get rid of Count Marivalle.
Cocinero    Siempre he querido ser cocinero profesional y tener mi propio negocio 
Chef   I’ve always wanted to be a professional chef and run my own business 
Cañonazo    Se celebra su centenario con un cañonazo desde el antiguo fuerte
Cannon fire    The centenary was celebrated with cannon fire from the old fort.
Pelotazo     Jugando un partido le dieron un pelotazo en toda la cara
Ball     A ball hit him right in the face when playing a match.
Dureza         Las imágenes eran de gran dureza y por eso se censuraron 
Harshness    Then harshness of the images led to their being censored. 
Mentiroso   Me asombra que sea tan mentiroso e insista en su versión
Liar   I am shocked by what a liar he is and how he sticks to his version of what happened.
Madurez   El hombre llega a la madurez alrededor de los veinte años
Maturity   A man reaches maturity around the age of twenty.
Coronilla       Los romanos se ponían en la coronilla un ramo de laurel
Crown       The Romans wore a laurel wreath on the crown of their heads.
Instrumental  El cirujano se quejó de que el instrumental era viejo
Equipment  The surgeon complained that the equipment was very old.
Andariego   El hijo de Laura ha salido muy andariego y no para de ir de aquí para allá 
Restless   Laura’s son has a very restless character and is always out and about. 
Corredizo   La cubierta tiene un cristal corredizo muy bonito
Sliding   The roof has a very pretty sliding glass door.
Pescador  El oficio de pescador tiene que ser muy duro
Fisherman  A fisherman lives a hard life.
Sencillez   Su pensamiento es de gran sencillez pese a lo profundo del mensaje
Simplicity   Her thinking exhibits great simplicity despite the depth of the message.
Novelista      Se consagró como gran novelista con su obra sobre la postmodernidad
Novelist      He achieved fame as a novelist with his work on postmodernism.
Mensajero  Recibí el paquete tarde así que le dije al mensajero que ya no lo quería 
Courier  I told the courier service I didn’t want the package because it arrived so late 

Page 47 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



For Peer Review

46

Olvidadizo  Me lo tendrás que repetir porque soy muy olvidadizo y no lo recordaré
Forgetful  You’ll have to repeat it because I’m very forgetful and won’t remember.
Desnudez  El concepto de desnudez se estudia en Bellas Artes
Nudity  The concept of nudity is studied in Fine Arts.
Pasaje       Le da miedo caminar por el pasaje que comunica las dos calles
Passage       He’s afraid to walk down the passage between the two streets.

Page 48 of 48

Scandinavian Journal of Psychology

Scandinavian Journal of Psychology

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60