Are you insulting me? The role of self-reference in the neural processing of pronouns. Research Master’s Psychology Thesis
University of Amsterdam Student: Mirela Zaneva Student number: 11119802 Supervisor: Dr. Marte Otten
Table of content Abstract ... page 3 Introduction ... page 3 - 5 Methods ... page 5 - 8 Participants ... page 5 - 6 Stimuli ... page 6 - 7 Procedure ... page 7 - 8 EEG Recording ... page 8 Data analyses ... page 7 - 8 Results ... page 8 - 16
Behavioral questionnaire ... page 8 - 10 EEG Results ... page 10 - 16
N1(100-150ms) ... page 11 P2 (200-300ms) ... page 12 N400 (300-500ms) ... page 13 - 14 Early LPP (900-1200ms) ... page 14 Late LPP (900-1200ms) ... page 15 Discussion ... page 16 - 19
Neural correlates of self-relevant and other-relevant pronouns... page 16 What may be driving the difference between our results and the self-relevance literature?
The reception of insults ... page 17 The importance of information about others ... page 17 - 18 Why did we find no evidence for a difference in processing of pronouns in insults and neutral statements? ... page 18 An alternative explanation: word frequency ... page 18 - 19 Conclusion ... page 19 -20 References ... page 21 - 23 Appendix ... page 24 - 34 Insulting sentences ... page 24 - 29 Neutral sentences ... page 29 - 34
Are you insulting me? The role of self-reference in the neural processing of pronouns. Abstract
The present study examines how the brain processes pronouns in an insulting context and how that processing differs between pronouns referring to the self and pronouns referring to others. We examined event-related potentials (ERPs) from 36 participants who read insulting and neutral sentences in Dutch. We found no evidence for a difference in the processing of pronouns in insults versus in neutral statements. However, our results indicated that other-relevant pronouns elicited higher late positivities in the left hemisphere and along the midline of the brain than self-relevant pronouns. We discuss this finding with relation to previous work in the self-relevance literature and the particularities of our methodology.
Introduction
Imagine hearing someone say “You are a fool!” Now imagine hearing instead “He is a fool!” These two sentences will likely be experienced very differently, yet they only vary in one word -- the pronoun that was used. The first sentence uses a self-relevant personal pronoun, you, and thus, the insult is likely to be experienced as harsher and more negative than the second sentence which, while also insulting, refers to another person. In this study, we explored the neural processing of personal pronouns in the context of insults, using event-related potentials (ERPs). We examined the differences between the neural responses to two types of pronouns: those referring to the self, and those referring to others.
Self-relevance plays an important role in how information is processed. As early as the 1950, researchers had realized that self-relevant information is privileged in our attention. Cherry (1953) discovered that a person’s name can capture her attention even when pronounced by an unattended speaker, a phenomenon known as the “cocktail party effect”. Indeed, one’s name captures one’s attention in the context of an unattended conversation or even in an unattended auditory channel (Moray, 1959). Multiple popular culture productions have capitalized on this effect and used it in movies and plays, demonstrating people’s innate desire to know the context in which they are referenced. Such desire makes sense evolutionarily, as it is important to know what other members of the social group have to say about you or want you to know.
It is especially crucial to know information about the self when the context is highly emotive, such as when an insult is involved. Emotions have been conceptualized as crucial for our species’ survival as they can be markers for threats and can guide approach and avoidance behaviors. Thus, it is important that our attentional resources are well attuned to keep track of whether emotive stimuli are related to us or not. According to appraisal theories of emotion, emotional cues are quickly appraised for their self-relevance (Scherer et al., 2001; Ellsworth and Scherer, 2003; Sander et al., 2005). Specifically, appraisal theories posit an automatic process of evaluating incoming input by firstly assessing self-relevance, and then the emotional nature of the content.
Recent work by Herbert and colleagues on self-relevance (2010, 2011) provides experimental evidence for these theoretical views. They demonstrated that words with negative valence (e.g.
“panic”) accompanied by the first-person possessive pronoun (“my”) are associated with higher late positive potentials (LPP) relative to words with negative valence accompanied by definite articles (“the”) and third-person pronouns (“his”). The authors argued that pronouns serve as markers designating the self-relevance of a given emotional stimulus for further processing. Indeed, extensive evidence, reviewed by Sui and Humphreys (2015), demonstrates that information, whose self-relevance is marked by pronouns, recruits additional cognitive processing and leads to better performance in perception, information integration and memory tasks relative to information which is marked as not relevant to the self.
While previous work, like the “cocktail effect” show that personal names, like pronouns, serve as cues for self-relevant information, philosophers of language distinguish pronouns and names as separate categories. Pronouns are categorized as indexicals, which means that the linguistic meaning of a pronouns such as ‘I’ is the same across context, but depending on who is speaking, the content of that pronoun changes (Kaplan, 1989). By contrast, most language philosophers posit that names are non-indexicals. Crucially, the essential indexical thesis (Castañeda, 1966) argues that self- reference differs between indexicals and non-indexicals. Castañeda reasons that personal pronouns are indispensable for capturing the expression of the self that one is immediately and most saliently aware of and thus cannot be substituted by one’s name. For instance, if asked who is writing this article, I would answer that I am, and not that Marte is or Mirela is. This differentiation between pronouns and names gives cause, then, to consider that self-relevance might be processed differently in cases when pronouns are used than in situations where personal names appear.
However, empirical evidence has disputed this theoretical distinction. Cross-cultural evidence from speakers of various languages shows that this dichotomization of the function of pronouns and names is not universal, but limited to Western languages (Jaszczolt, 2016). The psychological literature on self-relevance has also suggested that personal pronouns might be treated in similar ways to names in terms of information processing. Self-relevance elicited by visual and auditory representations of participants’ own personal names consistently resulted in larger LPPs (Berlad and Pratt, 1995; Folmer and Yingling, 1997; Perrin et al., 2005; Tacikowski and Nowicka, 2010), no differences in early components (before ~200ms) such as N1, P1, N170 and inconclusive results on later components (between 200 and 300ms) such as N2/N250 and P2 (e.g., Keyes et al., 2010; Sui et al., 2009; Tanaka et al., 2006). Turk and colleagues (2011) demonstrated that inducing self-relevance by using personal pronouns produces similar patterns of brain activity such as in the the above discussed literature with personal names. Hence, pronouns and names have both largely been treated as valid means of inducing self-relevance in the psychological literature.
Although little is known about the neural processing of personal pronouns in the context of insults in particular, psycholinguistic research has demonstrated that insults evoke additional processing relative to control stimuli. Otten, Mann, van Berkum and Jonas (2016) found that insults evoked a larger negativity between 300 and 400ms, followed by a short late positivity relative to compliments, suggesting an increase in LPP. These findings were consistent with the results of another recent ERP study by Struiksma, De Mulder, Spotorno, Basnakoca and Van Berkum (2014), implying an in-depth linguistic processing of insults.
The current study aims to build on these previous findings in the literature on self-relevance and the psycholinguistic literature regarding the processing of insults. We set off to examine the neural processing of personal pronouns in the context of insults and in particular, the differences in neural response to self-relevant versus other-relevant pronouns. We did this by presenting participants with a series of insulting and neutral statements in a computer task and recording event-related potentials (ERPs).
Methods Participants
38 participants (25 women, 13 men, mean age 23 years, range 19-36) whose first language is Dutch took part in the experiment. From these 38 participants, 2 were excluded from the analysis due to significant artifact presence in their data, thus leaving us with a sample of 36. The study was approved by the Ethical Review Board of the Department of Psychology at the University of Amsterdam. All participants read and signed an informed consent form. Further, we required that participants are not dyslexic, have normal or corrected to normal vision, have no neurological problems and are not currently on medication that would impede written language processing. Participants could not take part in the study if they have dreadlocks or very curly hair as this would impede data collection.
Stimuli
We constructed 60 insulting self-relevant sentences (e.g., “In photos you always look like the fattest one of the group.”), 60 insulting other-relevant sentences (e.g., “In photos she always looks like the fattest one of the group.”), 60 neutral self-relevant sentences (e.g., “In photos you always look like the tallest one of the group.”), and 60 neutral other-relevant sentences (e.g., “In photos she always looks like the tallest one of the group”). As illustrated in these examples, we constructed the stimuli so that they were completely identical up to the critical (insulting or neutral) word. If the critical insulting word was part of a phrase, we made sure the same number of words were used in the corresponding neutral statement. We also matched the number of words following each critical word. Thus, sentence length did not differ between the insulting and neutral sentences (M = 11.1 words, SD = 1.7 words). Self-relevant and other-relevant sentences differed only in the used pronoun, and hence had the exact same sentence length. All self-relevant sentences contained gender-neutral information in order for them to be suitable for all participants. Other-relevant sentences were evenly split between sentences in which “she” is the subject and sentences in which “he” is the subject. The position of pronouns was matched across insults and neutral sentences (Minsult = 4.6 words, SDinsult = 1.4, Mneutral = 4, SDneutral = 1.6, Mann-Whitney U = 349.5, p > .05) and was never beginning nor final. Based on the CELEX lexical frequency database (Baayen, Piepenbroek, & Gulikers, 1995), the other-relevant pronouns (“hij”, frequency = 470 924 per 1 million; “zijn”, frequency = 444 487; “zij”, frequency = 11 121; “haar”, frequency = 19 2007) used here are more frequent in the Dutch language than the self-relevant pronouns (“jij”, frequency = 21 624 per 1 million; “jouw”, frequency = 3830). In order to fully control the construction of the sentences, we were also mindful of position of the first explicitly insulting word in the sentence (i.e.
“fat” in the example above) and its neutral counterpart in the neutral sentences (i.e. “tall” in the example). Specifically, we matched the position of the insulting and neutral words across insults and neutral sentences (Minsult = 8.2 words, SDinsult = 1.63, Mneutral = 8.0, SDneutral = 1.86, Mann-Whitney U = 1656, p > .05) of sentences and was never final. The frequency of the critical words also did not differ between insults and neutral sentences (Minsult = 4803.7 occurrences per 1 million, SDinsult = 24535, Mneutral = 1367.6, SDneutral = 2251, Mann-Whitney U = 1619.5, p > .05). A complete list of the sentences is provided in the Appendix.
Procedure
The experiment had a 2 by 2 within-participants design: Self-Relevance Type (2 levels: self-relevant or other-relevant) by Emotional Content Type (2 levels: insult or neutral statement). The experimental stimuli were presented in a blocked manner. Each participant saw a total of two blocks: the first block contained 60 insults and the second block was comprised of 60 neutral statements. This type of block design ensured the integration of self-relevance within language processing. Participants had the opportunity to take small breaks after each round of 15 completed sentences, with opportunities to take slightly longer breaks after every 30 sentences. Participants were further informed that all sentences in a given block would be of the same type. Thus, participants knew what to expect and were aware if the sentence they would read was an insult or a neutral statement.
Within each block, self-relevant and other-relevant sentences were mixed. Sentence order was determined by lists to ensure that the self-relevance of individual insults and individual neutral statements was counterbalanced. That is, to ensure that half of the participants read a given insult with a self-relevant pronoun, and half of the participants read given insult with a other-relevant pronoun. Further, in order to assure that participants are aware of the self- and other-relevant condition, a photo of them was presented during self-relevant trials and neutral photos of strangers (either male or female) were presented during other-relevant trials (see Figure 1 for illustration). We took a photo of each participant in the lab, prior to starting the experiment. Participants were asked if they liked their photo and if they did not, we took a new one. We did this in order to prevent any negative emotions towards participants’ photos which could influence any potential effect in the experiment.
We also gaged participants’ appraisal of their emotions by prompting them to fill out a questionnaire at three different times: before the start of the experiment, at the halfway point (after the completion of the first block), and at the end of the experiment. Participants were given a list of possible emotional states (angry, humiliated, ashamed, worthless, outraged, cheerful, happy, grateful) and had to describe how they felt on a 1 to 7 likert scale, where 1 meant they absolutely did not feel the given emotion and 7 meant they absolutely did.
Sentences were presented on a monitor in front of the participants word for word via the Variable Serial Visual Presentation (VSVP) procedure (Otten, Nieuwland, & Van Berkum, 2007). This procedure aims to make visual presentation natural by varying the presentation time for each non-critical word depending on the word’s length. Non-critical word duration consisted of 240ms as a standard offset plus an additional 25ms per each letter. Non-critical words that were longer than
10 letters were presented for a fixed duration of 500ms. If a comma was present, we added 750ms to the word preceding the comma. The last words in each sentence received an additional 700ms presentation time. Critical words (insults and neutral) were never first or final words in any sentence and had a fixed presentation time of 425ms. Pronouns were similarly never presented at the start or end of a sentence and had a fixed presentation time of 375ms.
Figure 1. Illustration of sentence presentation in a self-relevant block. The participant first sees a start screen. They are informed they will read self-relevant sentences. A picture of themselves is presented. As words appear according to the Variable Serial Visual Presentation procedure, a smaller version of the participant’s photo remains in the corner of the screen.
EEG recording
Electrophysiological data was collected with BioSemi (Biosemi Inc., Amsterdam, The Netherlands) active-electrode system which includes 64+2 electrodes (Fp1, AF7, AF3, F1, F3, F5, F7, FT7, FC5, FC3, FC1, C1, C3, C5, T7, TP7, CP5, CP3, CP1, P1, P3, P5, P7, P9, PO7, PO3, O1, Iz, Oz, POz, Pz, CPz, Fpz, Fp2, AF8, AF4, Afz, Fz, F2, F4, F6, F8, FT8, FC6, FC4, FC2, FCz, Cz, C2, C4, C6, T8, TP8, CP6, CP4, CP2, P2, P4, P6, P8, P10, PO8, PO4, O2; CMS and DRL). We recorded blinks and eye-movements from two electrodes under and above the left eyes and an additional two electrodes on the inner and outer sides of the eye. For the purposes of offline re-referencing of the EEG signal, we placed an electrode on the left and right mastoids. The signal was amplified by a BioSemi ActiveTwo amplifier (-3dB at ~102 Hz low-pass, fully DC coupled) with a sample rate of 512 Hz.
We re-referenced the EEG signals offline to the average of the right and left mastoids and further applied offline high (0.01Hz) and low (60Hz) pass filters. Data were segmented into epochs which last from 200ms before stimulus onset until 1200ms after the onset of the critical word. We performed a baseline correction on the signals by subtracting the mean amplitude in the 150ms which precede word onset. A manual artifact rejection was performed in order to exclude any channels that recorded less than 18 out of 30 trial for each condition (i.e. 60% or more of the channel recording was not useable).
For each participant, artifact-free trials were averaged separately for the four conditions (self-relevant insults, other-relevant insults, self-relevant neutral statements, other-relevant neutral statements).
Data analyses
We performed repeated measures Analyses of Variance (ANOVA) for the N1 (100-150ms), P2 (200-300ms), N400 (300-500ms), early LPP (600-900ms) and late LPP (900-1200ms) time windows. In order to assess the effects of sentence type and pronoun, as well as any possible interaction with electrode position, the ERPs were evaluated with a repeated measures ANOVA crossing 4 factors: 1) Sentence type (insult vs. neutral statement), 2) Pronoun (self-relevant vs. other-relevant), 3) Hemisphere (left vs. right), and 4) Anteriority (anterior vs. posterior). Thus, the analysis involved four quadrants with respect to electrode position: 1) left-anterior quadrant comprising FP1, AF3, AF7, F7, F5, F3, F1, FC1, FC3, FC5, FT7; 2) right-anterior quadrant comprising FP2, AF4, AF8, F8, F6, F4, F2, FC2, FC4, FC6, FT8; 3) left-posterior quadrant comprising C1, C3, C5, T7, Cp1, Cp3, Cp5, TP7, P1, P3, P5, P7, P9, PO3, PO7, O1; and 4) right-posterior quadrant comprising C2, C4, C6, T8, Cp2, Cp4, Cp6, TP8, P2, P4, P6, P8, P10, Po4, Po8, O2. To assess the effects on the midline electrodes, Fpz, Afz, Fz, FCz, Cz CPz, Pz, POz, Oz, and Iz a separate repeated measures ANOVA was utilized, crossing three factors: 1) Sentence type (insult vs. neutral statement), 2) Pronoun (self-relevant vs. other-relevant), 3) Position. Interaction effects were examined with pairwise t-tests, Bonferroni corrected for multiple comparisons. A Greenhouse-Geisser adjustment was applied for F tests with more than one degree of freedom in the numerator, and reported uncorrected degrees of freedom and corrected P-values.
Results
Behavioral questionnaire
Since the data from the behavioral questionnaire were ordinal, we carried out the nonparametric Friedman test. For the positive emotions gratefulness, happiness, and cheerfulness we expected to see lower rates following the first block which contained insults. Further, we expected that participants would report higher levels of these positive emotions at the end of the experiment, after completing the second block, which contained neutral statements. As illustrated in Figure 2, this was indeed the case for all three positive emotions. The differences were significant for each emotion: cheerful: �2 = 8.860, N = 37, df =2 , p = 0.012, happy: �2 = 9.407 , N = 37, df = 2, p = 0.09, and
Figure 2. Means of reported scores per block for gratefulness, happiness, and cheerfulness in the behavioral questionnaire. The error bars represent one standard error above and below the mean. Note that the origin of the y-axis corresponds to an average rating of 3 and not 0. Positive emotions decreased following the completion of the first (insulting) block and then increased following the completion of the second (neutral) block.
Our questionnaire also examined how participants felt in regards to four other negative emotions: outrage, worthlessness, shame, humiliation, and anger. For these negative emotions we anticipated to observe an opposite trend to that of the positive emotions. We expected to see an increase in mean scores for negative emotions following the first block which contained insults. Following the second block, with neutral statements, we anticipated to see a decrease in the reported negative emotions.
Figure 3. Means of reported scores per block for outrage, worthlessness, and shame in the behavioral questionnaire. The error bars represent one standard error above and below the mean. Negative emotions increased following the completion of the first (insulting) block and then decreased following the completion of the second (neutral) block.
Indeed, as evident in Figures 3 and 4, this was the case for all seven negative emotions. Changes in all emotions were significant: angry: �2 = 15.6, N = 37, df = 2, p < .001, humiliated: �2 = 20.1, N
= 37, df = 2, p < .001, ashamed: �2 = 12.304, N = 37, df = , p = 0.002, worthless: �2 =12.667 , N
= 37 , df = , p < 0.002, outraged: �2 = 15.935, N = 37 , df = 2, p < .001.
Figure 4. Means of reported scores per block for humiliation and anger in the behavioral questionnaire. The error bars represent one standard error above and below the mean. Negative emotions increased following the completion of the first (insulting) block and then decreased following the completion of the second (neutral) block.
EEG Results N1 (100-150ms)
There was no main effect for sentence type (insult vs neutral statement) nor for pronouns on ERP between 100 and 150ms (Sentence type F(1) = .043, p = .836, �2 = .001; Pronoun F(1) = .84, p = .774, �2 = .002). However, there was a three-way interaction between hemisphere, sentence type and pronoun (F(1) = 8.023, p = .008, �2 = .186). Smaller N1s were observed in the left hemisphere than in the right hemisphere for self-relevant pronouns in insulting sentences but not in neutral sentences (t(35) = 2.82, p = .008, �2 = .186). Smaller N1s were also observed in the left hemisphere than in the right hemisphere for other-relevant pronouns in neutral statements but not for insulting sentences (t(35) = 2.82, p = .008, �2 = .186). These results are illustrated in Figures 5 and 6, respectively.
The results for the midline electrodes indicated an absence of main effect for sentence type and pronouns (Sentence type F(1) = .345, p = .561, �2 = .010; Pronoun F(1) = 1.007, p = .323, �2 = .029).
Figure 5. Grand averages recording for relevant pronouns. The black line denotes that the self-relevant pronoun was presented in an insulting context, whereas the red line signifies the context was neutral. Smaller N1s were observed in the left hemisphere than in the right hemisphere for self-relevant pronouns in insulting sentences but not in neutral sentences.
Figure 6. Grand averages recording for self-relevant pronouns. The black line denotes that the other-relevant pronoun was presented in an insulting context, whereas the red line signifies the context was neutral. Smaller N1s were observed in the left hemisphere than in the right hemisphere for other-relevant pronouns in neutral statements but not for insulting sentences.
P2 (200-300ms)
There was no main effect for sentence type (insult vs neutral statement) nor for pronouns on ERP between 100 and 150ms (Sentence type F(1) = .279, p = .601, �2 = .008; Pronoun F(1) = .354, p = .555, �2 = .01). However, there was an interaction between hemisphere and pronoun (F(1) = 6.62, p = .015, �2 = .159). Higher P2s were observed in the right hemisphere for self-relevant pronouns
and not in the left hemisphere (t(35) = 5.89, p < .001, �2 = .156). This is illustrated in Figure 7 which depicts grand averages and Figure 8 which contrasts representative electrodes from the left and right hemispheres.
The pattern of results in the midline electrodes also exhibited an absence of main effect for sentence type and pronouns (Sentence type F(1) = .671, p = .418, �2 = .019; Pronoun F(1) = .684, p = .414, �2 = .02).
Figure 7. Grand averages for self-relevant pronouns. Larger P2s were observed in the right hemisphere for self-relevant pronouns relative to the left hemisphere.
Figure 8. Recordings for self-relevant pronouns. The black line depicts the recording from F5, a frontal left hemisphere electrode, and the red line shows the recording from F6, a right hemisphere
frontal electrode. Larger P2s were observed in the right hemisphere for self-relevant pronouns relative to the left hemisphere.
N400 (300-500ms)
In the 300-500ms period, there was no main effect of either pronoun or sentence type (Sentence type F(1) = .034, p = .854, �2 = .001; Pronoun F(1) = 1.548, p = .222, �2 = .042). A statistically significant interaction between hemisphere and pronoun emerged (F(1) = 10.022, p = .003, �2 = 0.223), whereby there were differences in the N400 component for other-relevant pronouns between the two hemispheres (t(35) = 3.1, p = .004, �2 = .212). Specifically, other-relevant pronouns elicited smaller N400s in the left hemisphere in comparison to the right hemisphere. This is illustrated in Figure 9.
The pattern of results in the midline electrodes also exhibited an absence of main effect for sentence type and pronouns (Sentence type F(1) = .100, p = .754, �2 = .003 ; Pronoun F(1) = 1.698, p = .201, �2 = .048).
Figure 9. Grand averages for all other-relevant pronouns. other-relevant pronouns elicited larger N400s in the left hemisphere in comparison to the right hemisphere.
Early LPP (600-900ms)
There was no main effect for sentence type (insult vs neutral statement) nor for pronouns on ERP between 600 and 900ms in the hemisphere ANOVA (Sentence type F(1) = 1.223, p = .276, �2 = .034; Pronoun F(1) = 3.024, p = .091, �2 = .080). An interaction between hemisphere and pronoun trended towards significance (F(1) = 3.439, p = .072, �2 = 0.089). A visual inspection indicated that other-relevant pronouns tended to elicit larger early LPPs than self-relevant pronouns in the left hemisphere and not the right hemisphere (see Figure 10).
A main effect of pronoun was found in the midline section (F(1) = 4.787, p = .036, �2 = .123), such that other-relevant pronouns evoked larger early LPPs. This is illustrated in Figure 11 which shows the recording from a representative midline electrode, Fz.
No main effect was found for sentence type (F(1) = .119, p = .732, �2 = .003). Late LPP (900-1200ms)
The same pattern of results for the early LPP (600-900ms) was present in the later LPP period (900-1200ms). No main effect for sentence type (insult vs neutral statement) nor for pronouns was found in the hemisphere ANOVA (Sentence type F(1) = 2.01, p = .165, �2 = .054; Pronoun F(1) = 3.64, p = .065, �2 = .094). A significant interaction between hemisphere and pronoun, however, was apparent (F(1) = 6.7, p = .01, �2 = .161) whereby in the other-relevant pronouns elicited larger late LPPs than self-relevant pronouns in the left hemisphere but not the right hemisphere (t(35) = 2.53, p = .016, �2 = .154). This is illustrated in Figure 10.
In the midline region, the pattern again followed the one identified in the early LPP time epoch . A significant main effect of pronoun was present (F(1) = 7.62, p = .009, �2 = .183), whereby other-relevant pronouns elicited larger late LPPs. This is visible in the midline electrodes presented as part of Figure 9, as well as in Figure 11, which shows the recording from a representative midline electrode, Fz.
No main effect was found for sentence type (F(1) = .716, p = .403, �2 = .021).
Figure 10. Grand averages recordings. The black line denotes the self-relevant pronoun, whereas the red one shows the other-relevant pronoun. During the 900-1200ms time epoch, other-relevant pronouns elicited larger LPPs than self-relevant pronouns in the left hemisphere but not in the right hemisphere. During this same period (900-1200ms), a main effect of pronoun was present in the midline section, such that other-relevant pronouns elicited larger LPPs. During the 600-900ms time epoch, other-relevant pronouns tended to elicit larger LPPs than the self-relevant pronouns (but this did not reach significance).
Figure 11. Recording from the midline electrode Fz. The black line denotes the self-relevant pronoun, whereas the red one shows the other-relevant pronoun. A main effect of pronoun was found in the midline section during the 600-900ms time period, such that other-relevant pronouns elicited larger LPPs than self-relevant pronouns. In the 900-1200ms time period, we also found that other-relevant pronouns elicited larger LPPs in comparison to self-relevant pronouns.
Discussion
Neural correlates of self-relevant and other-relevant pronouns
Previous research has found evidence for a self-relevance effect on LPP, such that self-relevant words (e.g. names and pronouns) elicit a larger late positivity relative to non-self-relevant words (e.g. Herbert et al., 2010 & 2011). Our study did not replicate these findings. Instead, we found the opposite pattern -- other-relevant sentences elicited more positive ERPs in the period (600-1200) along the midline electrodes and in the left hemisphere. We discuss possible reasons for this finding in the following section.
Previous studies have largely found inconclusive results for a difference in neural processing between self-relevant and other-relevant words during earlier components (e.g., Keyes et al., 2010; Sui et al., 2009; Tanaka et al., 2006). The results of our study follow this line of evidence, having uncovered no main effect for pronoun type. However, we found that self-relevant pronouns elicited more positive ERPs in N1 in insulting sentences. It is likely that people’s attention may be captured by self-relevant pronouns, particularly when the context is negative.
What may be driving the difference between our results and the self-relevance literature? The reception of insults
We applied a blocked design in which one block comprised of only insulting sentences, and another block of only neutral statements. Within a given block, self-relevant and other-relevant sentences were interleaved. It is possible that participants dismissed the insults towards them as factually untrue, but treated the insults to others as more “interesting”, leading to the observed higher late positivities in response to other-relevant over self-relevant pronouns in the left hemisphere.
Indeed, several of our participants shared post-testing that they found the insults “funny”, as it was “not possible that such a thing would apply to” them. The rich literature on self-serving bias (Forsyth, 2008) suggests that people are quick to dismiss information that would negatively reflect on their image. Instead, as the name of the bias would suggest, people are eager to accept information which is self-serving. For instance, it has been empirically shown that participants are highly likely to reject the validity of negative feedback (Pyszczynski, Greenberg, & Holt, 1985) and unwilling to receive information about their poor performance and failings (Pyszczynski, Greenberg, & LaPrelle, 1985). The pervasiveness of the self-serving bias in a myriad of situations of people’s daily lives has been repeatedly established and it has been shown that the bias is strengthened in situations where the participant is experiencing “self-threat” (Campbell & Sedikides, 1999). That is, when people are faced with a situation where they may receive negative information about their intelligence, appearance, performance on a certain task or any other aspect of themselves that holds some importance to them, their behavior in line with the self-serving bias is amplified. Note that our insults constituted precisely such a “self-threat” situation, providing negative information about a participant’s intellect, appearance, etc. Thus, previous work on the self-serving bias suggests that our participants were likely to dismiss insulting statements, perhaps judging them as lies or as referring to some other person, which may explain why our results do not replicate previous findings for high LPPs in response to self-relevant pronouns.
The importance of information about others
Turning back to the idea that sentences about others were potentially treated as more interesting by our participants, paying more attention to information others, especially in situations when there is no insightful information about the self, could speak to an old evolutionary strategy. For instance, gossip, a long-existing form of gathering and dispersing information about others, has been treated as a prosocial behavior that aids relationship building, a way to gather reputational cues about other individuals in our environment, and even a method of observational learning which allows individuals to tailor their own behavior (Feinberg, Willer, Stellar, & Keltner, 2012; Baumeister, Zhang, & Vohs, 2004). Thus, our findings regarding a heightened lexical processing of other-relevant pronouns may point to the evolutionarily predisposed human interest in information about others. Further, gossip has been linked with the evolution of language. Dunbar (1998) argues that language evolved to meet our ancestors’ needs to exchange stories about others during social grooming. The left hemisphere is traditionally associated with language processing, and it has been explicitly linked with the evolution in language in humans and information exchange about others and the environment in primates (Corballis, 1991; Leavens, Taglialatela, & Hopkins, 2014).
Consequently, the observed higher late positivities in response to other-relevant pronouns in the left hemisphere are consistent with an account linking the observed results with the social exchange of information about others.
Why did we find no evidence for a difference in processing of pronouns in insults and neutral statements?
Our manipulation check indicated that on average, participants reported a shift towards slightly more negative emotions following insulting sentences. This is in line with the higher positivity we observed in early ERPs (100-150) in the left hemisphere for self-relevant insults, perhaps, corresponding to more heightened emotional/attentional processing (e.g. Chapman et al, 1978; e.g. Eason, 1981).
Other than this finding, our results did not indicate any main effect of sentence type nor an interaction between sentence type and pronoun. Given our discussion above for our participants’ reception of insults, it is possible that insults were treated as fabricated statements and were not received differently from their neutral counterparts. Thus, it is unlikely that there would be a difference in the processing of the two types of sentences.
An alternative explanation: word frequency
Our experiment contrasted self- and other-relevant pronouns. According to the CELEX lexical frequency database (Baayen, Piepenbroek, & Gulikers, 1995), the other-relevant pronouns are more frequent in the Dutch language than the self-relevant pronouns. Previous research has explored how different frequency words are processed. For instance, Fernández et al. (1998) contrasted ERPs for high and low frequency words. For the N400 component, high frequency words evoked more positive potentials. This is in line with our results which indicated that other-relevant pronouns (which are of higher frequency) elicited more positive ERPs during the 300-500ms period.
Further, Young & Rugg (1992) also investigated how different frequency words are processed. They examined the 500-900ms period and saw that higher frequency words evoked more positive ERPs. In our early LPP period (600-900ms) we saw that other-relevant pronouns tended to elicit more positive ERPs (although this did not reach significance), and in our later LPP period (900-1200) we observed a significant difference between other and self-relevant pronouns in the same direction, which was in line with Young and Rugg’s finding. Moreover, in their analysis of the 300-500ms epoch, they observed a significant frequency x hemisphere interaction, such that high frequency words were more positive in the left hemisphere than the right hemisphere. This aligns with our results during that same epoch which similarly indicated that other-relevant pronouns elicited more positive ERPs in the left hemisphere in comparison to the right hemisphere.
Keeping in mind how well our results map onto the findings from frequency research (as described above), it is possible that that the differences in pronoun frequencies underlie the findings from our study. Specifically, it may be that the frequency of pronouns, instead of the relevance type of the pronouns (self or other) drives the differences in ERPs. One way to investigate this possibility is by turning to other research on the topic of self-relevance. These studies have largely been carried out in English and German. Interestingly, the word frequencies of personal and possessive
pronouns in English follow an opposite trend from the one identified here for the Dutch language. The second person personal pronoun you (frequency = 10 000 per 1 million) is more common than both the third person personal pronouns he (frequency = 7637 per 1 million) and she (frequency = 3732 per 1 million) according to the SUBTLEX linguistic corpus (Van Heuven, Mandera, Keuleers, & Brysbaert, 2014). Like English, self-relevant pronouns in German are more frequent than other-relevant ones. According to the German version of SUBTLEX the pronoun you (“Ich”, frequency = 489637 per 1 million) is more frequent than the pronouns he (“Er”, frequency = 121 771 per 1 million) and the pronoun she (“Sie”, frequency = 366870 per 1 million). This gives cause to consider that perhaps word frequency is a confounding factor in previous studies that use English or German for their stimuli. Thus, it is plausible that the pattern of results uncovered by previous ERP studies with findings suggesting a more pronounced linguistic processing of self-relevant over other-relevant pronouns, may in fact reflect an effect of word frequency rather than self-relevance. If word frequency is indeed the driving force between the observed differential ERP processing, then this would explain the contradictory findings between our current experiment and other existing self-relevance experiments, particularly in the LPP time epochs. In turn, this could perhaps indicate that self-relevance is less important than previously considered, or at least not significant in a way that is measured by ERPs. A potential future direction that could parse out frequency and self-relevance effects could be to look at personal names. Previous research examining the effect of self-relevance via personal names has also established that a participant’s own name elicits higher LPPs relative to other personal names (Berlad & Pratt, 1995; Folmer & Yingling, 1997; Perrin et al., 2005; Tacikowski & Nowicka, 2010). However, perhaps, a participant has encountered their own name is more frequently than most other names (e.g. when friends and acquaintances address the person or when they have to introduce themselves, complete forms, etc.). Thus it may be hard to judge the frequency of one’s personal name, but records exist that indicate each personal name’s prevalence in the general population in a given time, which could be used as an index of frequency of the other-relevant personal names. The reaction a participant has to her own personal name can then be weighted against her reaction to more or less frequent personal names. More research is needed in order to assess whether the observed pattern of results both in our current study as well as in previous work is due to self-reference or rather, as we suggest, is driven by differences in word frequency.
Conclusion
In the present ERP study we investigated how the brain processes pronouns in an insulting versus a neutral context, and how that processing may be different when the pronouns refer to the self relative to when they refer to others. We found no evidence for a difference in the processing in pronouns in insults relative to in neutral statements. In contrast to previous work in the field, our results indicated higher late positivities in response to other-relevant pronouns as compared to self-relevant pronouns in the left hemisphere and along the midline of the brain. It has been largely established in the literature that self-relevant information, such as names, personal pronouns or nouns accompanied by possessive pronouns, elicit more pronounced late positivities. We discuss the reception of insults by our participants as a possible explanation for the lack of effect of emotional
context (i.e. insults vs. neutral statements) and propose word frequency as an alternative explanation for our finding for more pronounced lexical processing for the other-relevant compared to self-relevant pronouns. Our other-self-relevant pronouns, the third-person Dutch personal pronouns ‘he’ and ‘she’, are more frequent than our self-relevant pronoun, the second-person Dutch personal pronoun ‘you’. Our findings for higher late positivities for more frequent words are consistent with prior findings in the word frequency literature. What is more, when comparing word frequencies of self- and other-relevant pronouns in the languages in which previous studies examining brain processing of pronouns, we found that these results are also consistent with an explanation of word frequency. That is, in English and German it was the case that the self-relevant pronoun was more frequent than the other-relevant pronouns. Thus, it is possible that the observed higher late positivities in response to self-relevant pronouns were driven by their higher word frequency rather than by their self-relevance.
References
Baumeister, R. F., Zhang, L., & Vohs, K. D. (2004). Gossip as cultural learning. Review of general psychology, 8(2), 111.
Berlad, I., & Pratt, H. (1995). P300 in response to the subject's own name. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section, 96(5), 472-474.
Campbell, W. K., & Sedikides, C. (1999). Self-threat magnifies the self-serving bias: A meta-analytic integration. Review of general psychology, 3(1), 23-43.
Cherry, E. C. (1953). Some experiments on the recognition of speech, with one and with two ears. The Journal of the acoustical society of America, 25(5), 975-979.
Chapman, R.M., McCrary, J.W., Chapman, J.A., Bragdon, H.R. (1978). Brain responses related to semantic meaning. Brain and Language, 5 (2), 195–205.
Corballis MC (1991) The lopsided ape: evolution of the generative mind. Oxford University Press, New York.
Dunbar, R. (1998). Grooming, gossip, and the evolution of language. Harvard University Press.
Eason, R. G. (1981). Visual evoked potential correlates of early neural filtering during selective attention. Bulletin of the Psychonomic Society, 18(4), 203-206.
Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. Handbook of affective sciences, 572, V595.
Feinberg, M., Willer, R., Stellar, J., & Keltner, D. (2012). The virtues of gossip: reputational information sharing as prosocial behavior. Journal of personality and social psychology, 102(5), 1015.
Fernández, G., Weyerts, H., Tendolkar, I., Smid, H. G., Scholz, M., & Heinze, H. J. (1998). Event-related potentials of verbal encoding into episodic memory: dissociation between the effects of subsequent memory performance and distinctiveness. Psychophysiology, 35(6), 709-720. Folmer, R. L., & Yingling, C. D. (1997). Auditory P3 responses to name stimuli. Brain and language,
56(2), 306-311.
Forsyth, D. R. (2008). Self-serving bias. International encyclopedia of the social and behavioral sciences. (2) Herbert, C., Herbert, B. M., Ethofer, T., & Pauli, P. (2011). His or mine? The time course of self–
other discrimination in emotion processing. Social neuroscience, 6(3), 277-288.
Herbert, C., Pauli, P., & Herbert, B. M. (2010). Self-reference modulates the processing of emotional stimuli in the absence of explicit self-referential appraisal instructions. Social Cognitive and Affective Neuroscience, 6(5), 653-661.
Jaszczolt, K. M. (2016). Meaning in linguistic interaction: semantics, metasemantics, philosophy of language. Oxford University Press.
Jung, T. P., Makeig, S., Westerfield, M., Townsend, J., Courchesne, E., & Sejnowski, T. J. (2000). Removal of eye activity artifacts from visual event-related potentials in normal and clinical subjects. Clinical Neurophysiology, 111(10), 1745-1758.
Kaplan, D. (1989). Demonstratives’ in Themes from Kaplan, Almog, Perry Wettstein.
Keyes, H., Brady, N., Reilly, R. B., & Foxe, J. J. (2010). My face or yours? Event-related potential correlates of self-face processing. Brain and cognition, 72(2), 244-254.
Leavens, D. A., Taglialatela, J. P., & Hopkins, W. D. (2014). From grasping to grooming to gossip. In Evolution of Language: Proceedings of the 10th International Conference (EVOLANG10) (pp. 471-472).
Moray, N. (1959). Attention in dichotic listening: Affective cues and the influence of instructions. Quarterly journal of experimental psychology, 11(1), 56-60.
Otten, M., Mann, L., van Berkum, J. J., & Jonas, K. J. (2016). No laughing matter: How the presence of laughing witnesses changes the perception of insults. Social neuroscience, 1-12.
Otten, M., Nieuwland, M. S., & Van Berkum, J. J. (2007). Great expectations: Specific lexical anticipation influences the processing of spoken language. BMC neuroscience, 8(1), 89.
Pyszczynski, T., Greenberg, J., & Holt, K. (1985). Maintaining consistency between self-serving beliefs and available data: A bias in information evaluation. Personality and Social Psychology Bulletin, 11(2), 179-190.
Pyszczynski, T., Greenberg, J., & LaPrelle, J. (1985). Social comparison after success and failure: Biased search for information consistent with a self-serving conclusion. Journal of Experimental Social Psychology, 21(2), 195-211.
Perrin, F., Maquet, P., Peigneux, P., Ruby, P., Degueldre, C., Balteau, E., & Laureys, S. (2005). Neural mechanisms involved in the detection of our first name: a combined ERPs and PET study. Neuropsychologia, 43(1), 12-19.
Sander, D., Grandjean, D., & Scherer, K. R. (2005). A systems approach to appraisal mechanisms in emotion. Neural networks, 18(4), 317-352.
Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.). (2001). Appraisal processes in emotion: Theory, methods, research. Oxford University Press.
Sui, J., Liu, C. H., & Han, S. (2009). Cultural difference in neural mechanisms of self-recognition. Social Neuroscience, 4(5), 402-411.
Sui, J., & Humphreys, G. W. (2015). The integrative self: how self-reference integrates perception and memory. Trends in cognitive sciences, 19(12), 719-728.
Tacikowski, P., & Nowicka, A. (2010). Allocation of attention to self-name and self-face: An ERP study. Biological psychology, 84(2), 318-324.
Tanaka, J. W., Curran, T., Porterfield, A. L., & Collins, D. (2006). Activation of preexisting and acquired face representations: the N250 event-related potential as an index of face familiarity. Journal of Cognitive Neuroscience, 18(9), 1488-1497.
Struiksma, M. E., De Mulder, H. N. M., Spotorno, N., Basnakova, J., & Van Berkum, J. J. A. (2014). Insults hurt and they keep hurting, especially when they are about you! evidence from EEG and skin conductance measures. 21st Annual Meeting Cognitive Neuroscience Society, Boston, USA, April 5–8.
Turk, D. J., Van Bussel, K., Brebner, J. L., Toma, A. S., Krigolson, O., & Handy, T. C. (2011). When “it” becomes “mine”: Attentional biases triggered by object ownership. Journal of Cognitive Neuroscience, 23(12), 3725-3733.
Van Heuven, W. J., Mandera, P., Keuleers, E., & Brysbaert, M. (2014). SUBTLEX-UK: A new and improved word frequency database for British English. The Quarterly Journal of Experimental Psychology, 67(6), 1176-1190.
Young, M. P., & Rugg, M. D. (1992). Word frequency and multiple repetition as determinants of the modulation of event‐related potentials in a semantic classification task. Psychophysiology, 29(6), 664-676.
Appendix
Insulting sentences
Zodra jij een kamer binnenstapt, voelt iedereen gelijk afkeer vanwege jouw aanwezigheid. De scriptie die hij aan het schrijven is, is afgrijselijk slecht geschreven.
Het valt iedereen op dat jouw adem vandaag naar rotte vis ruikt.
De manier waarop hij communiceert, is kinderachtig vergeleken met die van anderen. In het verkeer fiets jij op een belachelijk idiote manier.
Op feestjes ben jij altijd te schijterig om met onbekenden te praten.
Het is algemeen bekend dat jouw presentaties enorm saai zijn om naar te luisteren. In sociale situaties is hij altijd het pispaaltje van de groep.
Uit haar levenskeuzes blijkt dat zij een afschuwelijk persoon is. Tijdens het uitgaan is hij altijd de ellendeling van de groep. Alle collegaas vinden dat zij een moodkiller is op werk.
Iedereen vraagt zich af waarom jij altijd op een vieze manier eet.
Elke keer dat zij met iemand praat, maakt zij een verschrikkelijke indruk. Iedereen zegt dat jij ontzettend vervelend bent tijdens meetings.
Iedereen denkt dat jij niet opgewassen bent tegen het volwassen leven.
In al jouw studiegroepen merken andere studenten gelijk op hoe dom en lui jij bent. Iedereen vindt dat haar politieke keuze achterlijk is.
Iedereen is het er mee eens dat jij een naar persoon bent.
Iedereen kan zien dat hij ontzettend belachelijk danst op feestjes. Vanwege haar arrogantie wordt zij nooit uitgenodigd voor feestjes. Tijdens het sporten komt zij als een onhandige sukkel over. Zodra hij aan het woord is, verveelt iedereen zich te pletter.
Na drie jaar sporten ben jij nog steeds een dik en afschuwelijk persoon. Op werk gedraagt hij zich als een tiran tegenover collegaas.
Naar jouw verjaardag gaan is een marteling voor iedereen.
Zelfs haar buren weten dat zij rond neukt wanneer zij op vakantie is. Wanneer hij met onbekenden praat, heeft hij erg slechte manieren. Op feestjes ben jij altijd de minst aantrekkelijke persoon.
Iedereen weet dat jij te dom bent om alledaagse zaken te begrijpen.
Afgelopen jaren is gebleken dat zij een arrogant en onbetrouwbaar persoon is. Op werk ben jij de meest onbekwame werknemer.
Op fotoos lijkt zij altijd de dikste van de groep. Op werk maak jij de meest idiote beslissingen.
Keer op keer blijk jij een ongelooflijk gemene vriend te zijn. Sinds januari loop jij de kantjes er van af op werk.
Niemand wil haar huis bezoeken, omdat zij een vies en smerig huis heeft.
Over het algemeen zorg jouw aanwezigheid ervoor dat mensen zich moedeloos voelen over hun leven.
Iedereen vindt dat hij er vandaag raar uitziet.
Iedereen weet dat in een relatie zitten jouw slechtste kant is.
Tijdens het zingen klinkt jouw stem alsof er nagels over een bord gaan. De ambities die jij hebt, zijn net zo doelloos als jouw leven.
Iedereen weet dat jij een onbenullige baan hebt.
Iedereen vindt dat jouw manier van denken erg bekrompen is. Veel studiegenoten vinden dat hij een enorm mislukt persoon is.
Iedereen weet dat zij zich als een harteloze misbaksel gedraagt tegenover haar ouders. Alle docenten op de universiteit zeggen dat hij een belachelijk luie student is.
Iedereen merkt op dat jij je altijd als een idioot gedraagt tijdens de les. De kleren die jij draagt, kunnen jouw gruwelijke lichaam niet verbergen.
De manier waarop zij dingen aanpakt, is afschuwelijk en onbedachtzaam om te zien. Wanneer hij nieuwe mensen ontmoet, zegt hij de meest idiote dingen.
Iedereen ziet dat jouw haar altijd ontzettend smerig is.
In het algemeen zijn jouw fooien aan het personeel aan de krenterige kant. Om zijn doelen te behalen, maakt hij misbruik van het vertrouwen van anderen. Collegaas zeggen dat hij een nietsnut is sinds maart.
Tijdens kerst gedraagt hij zich altijd ontzettend akelig tegenover zijn familie. In deze kamer wordt haar aanwezigheid niet gewaardeerd door anderen.
Op de universiteit wordt haar aanwezigheid als onaangenaam ervaren door anderen. Iedereen vindt dat jouw kledingstijl lachwekkend is tijdens zakelijke evenementen. Vanwege jouw schema wordt jij nooit uitgenodigd voor feestjes.
De manier waarop jij communiceert, is helder vergeleken met die van anderen. De ambities die hij heeft, staan centraal in zijn leven.
Iedereen vindt dat jij er vandaag ontspannen uitziet. Naar zijn verjaardag gaan, is een gewoonte voor iedereen.
In al haar studiegroepen merken andere studenten gelijk op hoe vlot zij op emails reageert. Op werk is zij de meest recent aangenomen werknemer.
Elke keer dat jij met iemand praat, maak jij graag onschuldige grapjes.
Iedereen weet dat jij je als een meegaand persoon opstelt tegenover jouw ouders. Op werk gedraag jij je zakelijk tegenover collegaas.
Iedereen weet dat hij een negen tot vijf baan heeft.
Collegaas zeggen dat jij naar de sportschool gaat sinds maart. Iedereen kan zien dat jij graag danst op feestjes.
Over het algemeen zorgt zijn eten ervoor dat niks verspild wordt. Op werk maakt hij erg veel beslissingen.
Zodra jij aan het woord bent, luistert iedereen aandachtig. De manier waarop jij dingen aanpakt, is logisch om te zien. Na drie jaar sporten is zij nog steeds een gemotiveerd persoon. Iedereen vindt dat haar manier van denken erg bedachtzaam is. Iedereen weet dat in een relatie zitten haar grootste wens is. Iedereen weet dat zij te druk is om alledaagse zaken bij te houden.
Zodra zij een kamer binnenstapt, voelt iedereen gelijk haar aanwezigheid aan. Iedereen vindt dat jouw politieke keuze recent gevormd is.
De scriptie die jij aan het schrijven bent, is in blauwe inkt geschreven. Het valt iedereen op dat zijn adem vandaag naar fruitige tandpasta ruikt. Wanneer jij nieuwe mensen ontmoet, vertel jij persoonlijke verhalen. Zelfs jouw buren weten dat jij vroeg slaapt wanneer jij op vakantie bent. Iedereen merkt op dat hij altijd bij het raam zit tijdens de les.
Uit jouw levenskeuzes blijkt dat jij een gelovig persoon bent.
Alle docenten op de universiteit zeggen dat jij een derdejaars student bent. In het verkeer fietst zij op een erg voorzichtige manier.
Keer op keer blijkt zij een dierenvriend te zijn.
Iedereen denkt dat zij uitkijkt naar het volwassen leven.
Iedereen vindt dat zijn kledingstijl gepast is tijdens zakelijke evenementen. In deze kamer is jouw aanwezigheid niet onopgemerkt gebleven.
Veel studiegenoten vinden dat jij een nuchter persoon bent. In sociale situaties ben jij altijd de jongste van de groep.
Niemand wil jouw huis bezoeken, omdat jij zo afgelegen woont. Iedereen ziet dat zijn haar altijd bedekt is.
Op de universiteit wordt jouw aanwezigheid verwacht door anderen. De kleren die zij draagt, kunnen haar tattoos niet verbergen.
Wanneer jij met onbekenden praat, vraag jij veel persoonlijke vragen. Iedereen is het er mee eens dat zij een praktisch persoon is.
Op feestjes is hij altijd bereid om met onbekenden te praten.
Het is algemeen bekend dat zijn presentaties aanbevolen zijn om naar te luisteren. Om jouw doelen te behalen, maak jij gebruik van adviezen van anderen.
Iedereen vraagt zich af waarom hij altijd op de Aziatische manier eet. Tijdens het uitgaan ben jij altijd de bob van de groep.
Tijdens het zingen klinkt haar stem als die van een ander persoon. Op fotoos lijk jij altijd de langste van de groep.
Sinds januari loopt hij stage op werk.
In het algemeen zijn zijn fooien aan het personeel aan de gemiddelde kant. Afgelopen jaren is gebleken dat jij een gezond persoon bent.
Tijdens kerst gedraag jij je altijd normaal tegenover je familie. Alle collegaas vinden dat jij kalm bent op werk.
In jouw relaties gedraag jij je als een normaal persoon. Iedereen zegt dat hij altijd oplet tijdens meetings. Op feestjes is zij altijd de minst beschonken persoon.
Over het algemeen zorgt haar aanwezigheid ervoor dat mensen zich bewust voelen van hun leven.
Neutral sentences
Iedereen merkt op dat hij zich altijd als een idioot gedraagt tijdens de les. Op werk gedraag jij je als een tiran tegenover collegaas.
Iedereen is het er mee eens dat zij een naar persoon is.
Op de universiteit wordt jouw aanwezigheid als onaangenaam ervaren door anderen. Veel studiegenoten vinden dat jij een enorm mislukt persoon bent.
Zodra jij aan het woord bent, verveelt iedereen zich te pletter. Wanneer jij met onbekenden praat, heb jij erg slechte manieren. Tijdens het uitgaan ben jij altijd de ellendeling van de groep. Iedereen vindt dat haar manier van denken erg bekrompen is. Tijdens het sporten kom jij als een onhandige sukkel over. In het verkeer fietst zij op een belachelijk idiote manier.
De manier waarop jij communiceert, is kinderachtig vergeleken met die van anderen. Iedereen vindt dat jouw politieke keuze achterlijk is.
Op werk is zij de meest onbekwame werknemer.
Iedereen vraagt zich af waarom hij altijd op een vieze manier eet.
Iedereen weet dat jij je als een harteloze misbaksel gedraagt tegenover je ouders. Niemand wil jouw huis bezoeken, omdat jij een vies en smerig huis hebt. In deze kamer wordt jouw aanwezigheid niet gewaardeerd door anderen. In sociale situaties ben jij altijd het pispaaltje van de groep.
Iedereen denkt dat zij niet opgewassen is tegen het volwassen leven. Tijdens het zingen klinkt haar stem alsof er nagels over een bord gaan. Over het algemeen zorgt zijn eten ervoor dat iedereen walging voelt. Iedereen weet dat in een relatie zitten haar slechtste eigenschap is.
Afgelopen jaren is gebleken dat jij een arrogant en onbetrouwbaar persoon bent. Het is algemeen bekend dat zijn presentaties enorm saai zijn om naar te luisteren.
Elke keer dat jij met iemand praat, maak jij een verschrikkelijke indruk. In het algemeen zijn zijn fooien aan het personeel aan de krenterige kant. Wanneer jij nieuwe mensen ontmoet, zeg jij de meest idiote dingen.
Iedereen vindt dat zijn kledingstijl lachwekkend is tijdens zakelijke evenementen. Om jouw doelen te behalen, maak jij misbruik van het vertrouwen van anderen. Iedereen zegt dat hij ontzettend vervelend is tijdens meetings.
Iedereen weet dat zij te dom is om alledaagse zaken te begrijpen. Collegaas zeggen dat jij een nietsnut bent sinds maart.
Het valt iedereen op dat zijn adem vandaag naar rotte vis ruikt.
Over het algemeen zorgt haar aanwezigheid ervoor dat mensen zich moedeloos voelen over hun leven.
De kleren die zij draagt, kunnen haar gruwelijke lichaam niet verbergen. Uit jouw levenskeuzes blijkt dat jij een afschuwelijk persoon bent. Vanwege jouw arrogantie wordt jij nooit uitgenodigd voor feestjes. Op feestjes is zij altijd de minst aantrekkelijke persoon.
De scriptie die jij aan het schrijven bent, is afgrijselijk slecht geschreven. Alle collegaas vinden dat jij een moodkiller bent op werk.
Op feestjes is hij altijd te schijterig om met onbekenden te praten. Iedereen kan zien dat jij ontzettend belachelijk danst op feestjes. Sinds januari loopt hij de kantjes er van af op werk.
Keer op keer blijkt zij een ongelooflijk gemene vriendin te zijn.
Alle docenten op de universiteit zeggen dat jij een belachelijk luie student bent. Naar zijn verjaardag gaan is een marteling voor iedereen.
Op werk maakt hij de meest idiote beslissingen.
De ambities die hij heeft, zijn net zo doelloos als zijn leven.
De manier waarop jij dingen aanpakt, is afschuwelijk en onbedachtzaam om te zien. Tijdens kerst gedraag jij je altijd ontzettend akelig tegenover je familie.
Op fotoos lijk jij altijd de dikste van de groep.
In al jouw relaties gedraag jij je als een slecht persoon.
Zodra zij een kamer binnenstapt, voelt iedereen gelijk afkeer vanwege haar aanwezigheid. Zelfs jouw buren weten dat jij rond neukt wanneer jij op vakantie bent.
Iedereen vindt dat jij er vandaag raar uitziet. Iedereen weet dat hij een onbenullige baan heeft.
In al haar studiegroepen merken andere studenten gelijk op hoe dom en lui zij is. Iedereen vindt dat jouw kledingstijl gepast is tijdens zakelijke evenementen. Iedereen merkt op dat jij altijd bij het raam zit tijdens de les.
Naar jouw verjaardag gaan, is een gewoonte voor iedereen. Iedereen denkt dat jij uitkijkt naar het volwassen leven.
Het valt iedereen op dat jouw adem vandaag naar fruitige tandpasta ruikt. Op de universiteit wordt jouw aanwezigheid verwacht door anderen. Na drie jaar sporten ben jij nog steeds een gemotiveerd persoon. In deze kamer is haar aanwezigheid niet onopgemerkt gebleven.
Wanneer hij met onbekenden praat, vraagt hij veel persoonlijke vragen. Iedereen is het er mee eens dat jij een praktisch persoon bent.
Alle collegaas vinden dat zij kalm is op werk.
Om zijn doelen te behalen, maakt hij gebruik van adviezen van anderen. Tijdens kerst gedraagt hij zich altijd normaal tegenover zijn familie. Iedereen weet dat jij een negen tot vijf baan hebt.
Zodra hij aan het woord is, luistert iedereen aandachtig.
Iedereen ziet dat jouw haar altijd bedekt is.
In sociale situaties is hij altijd de jongste van de groep.
Vanwege haar schema wordt zij nooit uitgenodigd voor feestjes. De manier waarop zij dingen aanpakt, is logisch om te zien. Op werk ben jij de meest recent aangenomen werknemer. Op feestjes ben jij altijd de minst beschonken persoon.
Elke keer dat zij met iemand praat, maakt zij graag onschuldige grapjes. In het verkeer fiets jij op een erg voorzichtige manier.
Wanneer hij nieuwe mensen ontmoet, vertelt hij persoonlijke verhalen.
Het is algemeen bekend dat jouw presentaties aanbevolen zijn om naar te luisteren. De ambities die jij hebt, staan centraal in jouw leven.
De kleren die jij draagt, kunnen jouw tattoos niet verbergen.
In het algemeen zijn jouw fooien aan het personeel aan de gemiddelde kant. Zelfs haar buren weten dat zij vroeg slaapt wanneer zij op vakantie is. Keer op keer blijk jij een dierenvriend te zijn.
Iedereen vindt dat haar politieke keuze recent gevormd is. Iedereen vindt dat hij er vandaag ontspannen uitziet. Veel studiegenoten vinden dat hij een nuchter persoon is. Tijdens het uitgaan is hij altijd de bob van de groep. Op werk gedraag jij je zakelijk tegenover collegaas.
Tijdens het zingen klinkt jouw stem als die van een ander persoon. Sinds januari loop jij stage op werk.
Op werk maak jij erg veel beslissingen.
Tijdens het sporten komt zij als een geconcentreerd persoon over. Alle docenten op de universiteit zeggen dat hij een derdejaars student is.
Afgelopen jaren is gebleken dat zij een gezond persoon is.
In al jouw studiegroepen merken andere studenten gelijk op hoe vlot jij op emails reageert. Iedereen weet dat jij te druk bent om alledaagse zaken bij te houden.
Iedereen kan zien dat hij graag danst op feestjes.
Over het algemeen zorgt jouw eten ervoor dat niks verspild wordt. De scriptie die hij aan het schrijven is, is in blauwe inkt geschreven. Iedereen vindt dat jouw manier van denken erg bedachtzaam is. Collegaas zeggen dat hij naar de sportschool gaat sinds maart. Op feestjes ben jij altijd bereid om met onbekenden te praten. Iedereen weet dat in een relatie zitten jouw grootste wens is. Uit haar levenskeuzes blijkt dat zij een gelovig persoon is.
Iedereen weet dat zij zich als een meegaand persoon opstelt tegenover haar ouders. De manier waarop hij communiceert, is helder vergeleken met die van anderen.
Over het algemeen zorgt jouw aanwezigheid ervoor dat mensen zich bewust voelen van hun leven. Op fotoos lijkt zij altijd de langste van de groep.
In al haar relaties gedraagt zij zich als een normaal persoon. Niemand wil haar huis bezoeken, omdat zij zo afgelegen woont.
Zodra jij een kamer binnenstapt, voelt iedereen gelijk jouw aanwezigheid aan. Iedereen zegt dat jij altijd oplet tijdens meetings.
