The emotion-gender stereotype can cause perceptual
illusions and maybe short-term memory illusions.
Name student: Bente Olsthoorn
Student number: 12166987
Name supervisor: Marte Otten
Submission date: 22-01-2021
Abstract
The emotion-gender stereotype is one of the most common and influential
stereotypes there is. Women are often seen as more emotional and are expected to show sadness and happiness and less anger. Men, on the other hand, are expected not to show sadness but to show anger. Earlier research showed that expectations about the world can cause illusions in the short-term memory. The experiment in this paper tests whether the emotion-gender stereotype causes short-term memory illusions. Before the real experiment, an implicit association test was done to make sure the stereotype exists in our subject pool. In the real experiment, participants were instructed to remember a memory display, after which one of the pictures in this display was cued. The memory display consisted of 4 pictures, 2 male and 2 females, of which 2 were sad-looking, and 2 were angry looking. Participants made more mistakes when the cued picture was in line with the stereotype than when it was not in line with the stereotype. However, the difference in the number of
mistakes between the incongruent and congruent target did not increase when the participants had to remember the picture for a longer time. This means that it can not be concluded that the emotion-gender stereotype causes short-term memory illusions. However, the stereotype can cause perceptual illusions.
Introduction
There are many stereotypes involved around gender, but one of the most common gender stereotypes is probably the gender-emotion stereotype. “Women are too emotional for leadership” and “It is not manly to cry” are common sentences in our society and
originate from the ideas we have about what men and women are supposed to be doing and feeling. Research shows that people think that women experience and express almost every emotion more than men (Fabes & Martin, 1991). Only anger and pride were thought to be expressed and experienced more by men (Plant, Hyde, Keltner & Devine, 2000). This paper tests whether this emotion-gender stereotype can cause illusions in the short-term memory. The hypothesis is that emotion-gender stereotypes can cause short-term memory illusions. This means that, when these illusions do occur, they can also be saved into our long-term memory, which can cause a wrong image of someone. This can have significant effects and could be a reason for the glass ceiling, pay gap, and other gender discriminations (Brescoll, 2016).
The idea that men express more anger and pride, and women show more of the other emotions, shows in our interpretation of ambiguous pictures (Plant et al., 2000). Women were interpreted as sadder and less angry than the pictures of men, and vice-versa. There were two different ambiguous pictures: one where the upper half of the face was sad and the lower half angry, and one where these emotions were turned around (upper angry, lower sad). The participants rated four emotions (sadness, sympathy, anger, and contempt) on the amount they were expressed by the picture on a scale from 1-7 (1 is no emotion, 7 is extreme emotion). Even unambiguous pictures of angry women were interpreted as sadder and less angry than the unambiguous pictures of angry men in this experiment.
Hess, Adams, Grammer & Kleck (2009) showed their participants androgynous faces showing either an angry, happy, or fearful emotion. The research showed that the
participants associated the angry face more with men than with women. The other emotions were more likely to be associated with women. In the second study Hess et al. (2009) did, the participants saw male and female faces, showing neutral, angry, sad, happy, and fearful emotions. The participants had to indicate as fast as possible if the face they saw was male or female. The results showed that the participants were slower when indicating an angry woman than a woman showing any other emotion. So, Hess et al. showed that anger is associated more with men.
The research of Parmley & Cunningham (2014) showed that sad expressions were identified more quickly in women than in men. Besides, angry expressions were identified more accurately in men than in women. This was researched by showing participants morphed videos of men and women going from a neutral expression to a sad or angry expression. The participants had to indicate when they first saw an emotion appear and when they were sure of the emotion that was shown by the target. At both times, they had to indicate which emotion they thought they saw. Female participants identified the sad expressions of women earlier than of men. Additionally, the angry emotion shown by men was identified more accurately than the angry emotion shown by women.
This stereotype is already visible in children of a young age. When shown a face with ambiguous gender, children tended to label the sad, happy, and fearful faces as female and the angry faces as male (Birnbaum, Nosanchuk & Croll, 1980). Condry & Condry (1976) showed that people tend to rate emotions shown by children differently, depending on their gender. This means that gender is important in labeling different emotions.
In earlier research, it has been shown that gender stereotypes can affect our long-term memory and thus can cause long-long-term memory illusions (Lenton, Blair & Hastie, 2001). In this research, the participants were instructed to learn a list of words. This list was
composed of smaller lists. One of the smaller lists included only words that were either stereotypically male roles or stereotypically female roles. After the participants had learned the words, they got a recognition test with studied and non-studied words. The non-studied words included roles and traits that are stereotypically male and stereotypically female. The participants had to indicate which words they had learned earlier and which words were new. The results showed that the participants who learned stereotypically female words were more likely to falsely recognize other stereotypically female words. The same results were found for stereotypically male words.
Pinto, Seth & Otten (in prep) showed that people can have illusions in the short-term memory too, and that expectations can cause these illusions. The participants were shown six letters in a circle. Half of the letters were normal, and half of them were mirrored. One of these letters was cued and had to be remembered by the participant. After the cue, an irrelevant display was shown. The participants had to indicate which letter they saw, and whether it was mirrored or not, and after that, indicate how confident they were about their answer. This is necessary because guesses could be filtered out that way. Results show that the participants more often made the mistake of seeing the mirrored letter but
remembering the normal letter than vice versa. This mistake is made because it is in accordance with our knowledge of the world, and happens in a period of 3 seconds, which makes it a short-term memory illusion.
These false memories can occur in both the short and long-term memory. A false memory means that something that has to be remembered is misremembered. Most of the false memories arise through the interference of past or future events (Loftus & Pickrell, 1995). Internal expectations are one of the things that can interfere with our memory. Internal expectations are expectations we have about how the world works and the
knowledge we already have. Internal expectations can also include stereotypes, for example, the emotion-gender stereotype. In this case, people can adjust their memories to the
emotions they expect men and women to show.
Besides memory illusions, perceptual illusions also exist. To make sure the illusion is a short-term memory illusion instead of a perceptual illusion, one has to look if there is an increase in mistakes made over time. If this is the case, it is a short-term memory illusion. If this is not the case, it is a perceptual illusion because perceptual mistakes can only be made when the participant sees the thing he/she has to remember (Flegal, Atkins & Reuter-Lorenz, 2010).
In this paper, two experiments were done. First, an implicit association test (IAT) is carried out with pictures of men and women and sad and angry words, to make sure the emotion-gender stereotype exists. After we made sure the stereotype exists, the actual research starts. This research is based on the research that was done by Pinto et al. (in prep). In this experiment, the participant will see several pictures of men and women showing different emotions and has to remember the face and the emotion of one of the pictures. After the participant had chosen the picture he/she thought was right, the participant had to indicate their confidence about the answer (see Figure 4 in the methods section).
The expectation is that people make more mistakes that are in line with their internal expectations about the emotions that are shown by men or women. If this is the case, people will say with high confidence that they saw a sad woman, even if they saw an angry woman. This mistake will be made more often than vice versa. The same mistake, which is in line with the gender-emotion stereotypes, will also be made with men. The subjects will then say, with high confidence, that they saw an angry man, even though they saw a sad man. The number of mistakes will also go up when the participants have to remember the pictures for longer.
Implicit association test (pre-test)
To make sure that the emotion-gender stereotype exists, we first did an implicit association test (IAT). After this, the real experiment was executed.
Method
Participants
Sixty-seven students participated in this experiment, ranging from 16-49 years old (17 males, 50 females, average age: 20,2), most of them from the University of Amsterdam. 1 person was excluded from the data due to excessive speed. Only Dutch students were included because they are very familiar with the words and will categorize them better. All participants gave their written informed consent to participate in this experiment, which was approved by the ethics review board of the Psychology Department at the University of Amsterdam. The students participated for student credits.
Stimuli
The stimuli for this IAT were photos and words. The pictures used in this IAT came from the Radboud Faces Database (RaFD) (Langner et al., 2010). The pictures showed the faces of 6 Caucasian males and 6 Caucasian females, all with a neutral expression (Figure 1). The pictures were 681x1024 pixels. Also included in this IAT were 6 Dutch words with an angry meaning and 6 Dutch words with a sad meaning (Table 1).
Angry words Sad words Boos Woedend Kwaad Geïrriteerd Razend Agressief Verdrietig Neerslachtig Triest Droevig Somber Treurig
Table 1: Words used in the IAT.
Figure 1: Examples of pictures used in the IAT.
Procedure
The task was made with IATgen.org (Carpenter et al., in press). The different stimuli were loaded into this program, which made an IAT that could be executed in Qualtrics.com (Qualtrics, Provo, UT).
The IAT exists of 7 different rounds (Figure 2). In the first two rounds, which are for practice and consist of 20 trials each, the participants would see two categories, one on the left side of the screen and one on the right side of the screen. In this case, these categories were angry and sad or male and female. The participant would see a picture or a word (never together, always in different trials) in the middle. The participant will have to quickly and correctly categorize the picture or word in one of the categories by pressing the E (if it fits the left category) or the I (if it fits the right category). In the third and fourth round, the participants will see two categories on the screen’s left side (for example, male and angry) and two categories on the right side (for example, female and sad). The participant will see both pictures and words and will have to correctly categorize these words and pictures, again by pressing the E or I. The third and fourth rounds are the same, only the third round is a practice round and consists of 20 trials, and the fourth round consists of 40 trials. Both the third and the fourth round are used in the data analysis.
The first round is repeated in the fifth round, but the categories of this stage are swapped. This round also consists of 40 trials. These categories stay swapped in the sixth and seventh round, which are equal to the third and fourth round, only one of the categories at each side is swapped (for example, on the left, it will now say female and angry, and on the right male and sad). The sixth round consists of 20 trials, and the seventh round of 40 trials. The sixth and seventh rounds are used in the data analysis.
The participants could take a break in between the blocks and were asked after the experiment if they wanted to exclude their data from the analysis.
Figure 2: The different rounds of the IAT.
Analysis
The analysis of the data is done by IATgen.org (Carpenter et al., in press). Trials were excluded from the analysis if the participant took more than 10 seconds to answer. When more than 10% of the trials are answered in 300 ms., all the participants’ trials were excluded (Greenwald et al., 2003). Rounds 3, 4, 6, and 7 are included in the analysis. In the third and fourth rounds, the categories on each side are congruent with the stereotype. In the sixth and seventh, the categories are incongruent with the stereotype (Bhati, 2020).
The D-score algorithm is used to score the IAT. The mean reaction times of rounds 4 and 7 and rounds 3 and 6 are subtracted and divided by the standard deviation (SD) to get the D-score (Greenwald, Nosek & Binaji, 2003). A positive D-score indicates a preference for men and anger together and women and sadness together. A negative D-score indicates a preference of men and sadness, and women and anger together. The D-score can be minimum -2 and maximum +2 (Bhati, 2020).
The D-score is analyzed by a t-test, which will indicate whether the D-score is significantly different from zero or not. Besides the t-test, the reliability (estimated internal
consistency of IAT, based on split-half with Spearman-Brown correction) will also be
calculated. A reliability of 80% is best. However, most scales use a cut-off percentage of 70% to indicate whether the IAT is reliable (Bhati, 2020).
Results
The IAT is scored with a D-score algorithm by IATgen.com (Carpenter et al., in press). The mean D-score that was found was 0,28937 (sd = 0,34669) (t (65) = 6.78083, p < 0.00001). The reliability of the IAT is 79,359%. If de d-score differs significantly from 0, people respond significantly faster to one pair of categories than the other.
Discussion
Because the D-score of the IAT is positive and the reliability is close to 80%, it can be concluded that in this sample, people implicitly associate anger more with men than with women and associate sadness more with women than with men. The IAT is a powerful measure of implicit associations and stereotypes (Rudman, Greenwald, Mellott & Schwartz, 1999). However, the participants that did this IAT are not comparable with Dutch society as a whole, because the participants are all students. Nevertheless, because the participants for the gender memory test are also all students, it can be concluded that the emotion-gender stereotype exists in our population of participants and that the real experiment could be continued.
Gender Memory Test
MethodParticipants
Participants were 36 students, ranging from 18-49 years old (7 males, 29 females, mean age: 20,4), most of them from the University of Amsterdam. One participant was excluded due to not understanding the experiment. All participants gave their written
informed consent to participate in this experiment, which was approved by the ethics review board of the Psychology Department at the University of Amsterdam. The students
participated for research credits. Stimuli
The pictures used in the gender memory test come from the Radboud Faces
Database (RaFD) (Langner et al., 2010). The pictures included 16 different faces of Caucasian men and 16 different faces of Caucasian women, both looking angry and sad. In total, there were 64 pictures (see figure 3). The pictures were 681x1024 pixels.
Figure 3: Examples of pictures used in the gender memory test.
Procedure
This experiment was made and run with Neurotask.com (Neurotask online testing, 2020).
The participants joined a Zoom meeting with the test leader before the experiment started. In this meeting, the test leader made sure the instructions were clear and everything worked. After some practice trials, the test leader would leave the break-out room to let the participant do the experiment in private.
Every trial looked the same (Figure 4). The trials started with a fixation cross in the middle of the screen for 2 seconds. Then, the memory display followed. The memory display showed four faces; half were male, and the other half were female. Half of the faces showed an angry emotion, and half of the faces a sad emotion. This memory display is visible for 0,8 seconds. After half of these memory displays, the fixation dot appeared for 0,5 seconds. After the other half, the fixation cross appeared for 3 seconds. Following this fixation cross, a cue appeared on the place of one of the pictures for 0,1 seconds. The picture that was located on this place was the picture the participants had to remember and report. When the cue was gone, the participant saw a response display with eight faces with four different identities. Two of the identities in the response display were also shown in the memory display, the target identity and another identity. The other two identities that were shown in the response display were not shown in the memory display. There were two pictures of every identity in the response display, one showing a sad emotion and one showing an angry emotion. Also, half of the identities were male, and half were female. The participant had to choose which of the faces he/she saw by pressing key 1-8. After this, the participant had to indicate his/her confidence about their answer on a scale from 1 to 4, where 1 is a guess, and 4 is certain.
Before the experiment starts, the participant had to go through an example and a practice round. These rounds were made in Qualtrics.com (Qualtrics, Provo, UT). In the example round, the participants saw an overview of the different stages of the experiment and gave an answer based on those pictures. The practice trials looked the same as the real trials, except the time that the memory display was visible got shorter with every trial, until it was 0,8 seconds. There were three example trials and seven practice trials. The real trials existed of 16 blocks, with in each block 24 trials.
In the example and practice trials, the participants received feedback after each trial. In the real trials, the participants received their percentage of right answers after each block. The entire experiment took about one and a half hour.
Figure 4: Different stages of one trial of the Gender Memory Test.
Analysis
All the example and practice trials are excluded from the analysis. Only trials where the participant indicated a confidence level of 3 or 4 were included in the analysis. This ensures that the false alarms are not a product of guesses but of real illusions in the participant’s memory. To look at the number of false alarms that occurred, a repeated measures (RM) ANOVA is carried out. The RM ANOVA hastwo factors, memory delay (interval of 150 milliseconds or 3 seconds) and the stereotype congruity of the target
(congruent (con) or incongruent (inc) with the stereotype). Only false alarms (the participant has chosen the right identity but with the wrong emotion) are the dependent variable. The analysis of the data will be done in JASP (JASP Team, 2020).
When significant results are found within the ANOVA, a post-hoc test is done. This post-hoc test consists of paired t-tests corrected by the Bonferroni-correction because of the multiple comparisons. If the ANOVA provides insignificant results, a Bayesian analysis will check whether the insignificance indicates that the null hypothesis is true (Rouder, Speckman, Sun, Morey & Iverson, 2009). The Bayes factor provided by this analysis will indicate how strong the evidence is for the null or alternative hypothesis (Bayes Factor = BF. BF < 0.1: strong evidence for the null hypothesis. 0.1 - 0.33: substantial evidence for the null hypothesis. 0.33 - 1: anecdotal evidence for the null hypothesis. 1 – 3: anecdotal evidence for the alternative hypothesis, 3 – 10: substantial evidence for the alternative hypothesis, >10: strong evidence for the alternative hypothesis (Jeffreys, 1961, Wagenmakers, Wetzels, Borsboom, & Van Der Maas, 2011)).
The hypothesis would be confirmed if there are more false alarms in the trials with incongruent targets than with congruent targets, resulting in a main effect of congruity. A main effect of interval also has to be visible, where the trials with a longer interval have more false alarms than the trials with a shorter interval. Besides this, the difference in the number of false alarms between the incongruent and congruent trials has to go up with the longer interval, resulting in an interaction effect.
Results
The two-way repeated measures ANOVA showed a significant difference in the number of false alarms between congruent targets (mean = 15,4 %, sd = 0,093) and
incongruent targets (mean = 18,45%, sd = 0,113, F (1,35) = 7,725, p = 0,009, η2 = 0,052). The
difference in the number of false alarms is also significant between the interval of 150 ms (mean = 15%, sd = 0,097) and 3000 ms (mean = 18,85%, sd = 0,109, F (1,35) = 9,376, p = 0,004, η2 = 0,084). The interaction between congruity and interval did not show a significant
difference between the number of false alarms (F (1,35) = 0,285, p = 0,597, η2 = 0,003, BF =
0,204). The difference in false alarms between congruent and incongruent trials is 2,4% for the short interval and 3,7% for the long interval. The Bayesian analysis that is done, indicates that there is substantial evidence for the null hypothesis. These results are also shown in figure 5.
Because the interaction between congruity and interval is not significant, it is not needed to do a post-hoc test.
Figure 5: Percentage of high confidence false alarms (FA) for short and long memory delay (150 or 3000 ms) and congruent and incongruent targets. The error bars report the standard error.
0.00% 5.00% 10.00% 15.00% 20.00% 25.00% 150 ms 3000 ms Perc en ta ge FA congruent incongruent
Discussion
In summary, the results showed that more false alarms were made when the target was incongruent with the emotion-gender stereotype than when it was congruent. A longer memory delay also caused more false alarms. The difference in the number of false alarms between congruent and incongruent targets did not increase with the longer interval. From these results, it seems that the emotion-gender stereotype does influence what people report to have seen. However, it can not be concluded that the emotion-gender stereotype affects the short-term memory.
The results also showed that targets that are incongruent with the emotion-gender stereotype caused more false alarms than congruent targets. However, the results did not show an effect on the interaction between target type and interval on the relative
occurrence of memory illusions. An explanation for this finding could be that we observed perceptual illusions. With perceptual illusions, the mistake can only be made when looking at the pictures. After that, the illusion is there and thus can not arise over time. The number of illusions will stay the same over time (Flegal et al., 2010). This is different from short-term memory illusions. These illusions can only arise after the participant looked at the pictures. The longer the participant has to remember the pictures, the harder it gets, and the more illusions arise (Pinto et al., in prep). However, the difference between the number of false alarms of the incongruent and congruent target was slightly bigger with the longer interval (3,7%) compared to the short interval (2,4%). This difference was not big enough to conclude that the stereotype causes short-term memory illusions. Nevertheless, the research of Pinto et al. (in prep) did show that expectations, based on real-world knowledge, can cause short-term memory illusions. Besides, Lenton et al. (2001) showed that social prejudices could cause false memories of words that are in line with the emotion-gender stereotype. It is possible that there could be an interaction effect between congruity and interval if more participants participated in this study. If this is the case, then the emotion-gender stereotype does affect the short-term memory.
Furthermore, the results showed that a longer interval caused more false alarms than a shorter interval. This is because the longer interval is more challenging in all cases, no matter the target. After all, the participants have to remember the faces longer. Participants indicated that they found it hard to remember all the faces, so it is even more challenging when the participants have to remember the faces for a longer time.
Another reason that could have influenced the results is that both a confidence level of 3 and 4 were included. Some participants indicated that they pressed 4 when they were entirely sure and pressed 3 when they were sure about the picture’s identity, but not sure about the emotion. This means that we also included some trials where the participants guessed which picture was the right picture. When this is the case, false alarms are not classified as an illusion, either perceptual or short-term memory. With more participants, it will be possible only to include a confidence level of 4.
That the emotion-gender stereotype influences the perception, and maybe the short-term memory, can have some implications, especially work-related. Women who show more anger at work get adverse reactions and lower ratings, simply because this emotion is not expected to be shown by a woman (Heilman & Okimoto, 2007; Lewis, 2000). However,
penalized. Besides, the stereotypically female emotions, such as sadness, are not ‘dominant’ emotions and thus are not perceived as emotions that positively influence leadership. Anger and pride are seen as ‘dominant’ emotions, and because men are expected to show these emotions, they are also expected to be a better leader (Brescoll, 2016). The emotion-gender stereotype influences people’s perception and can thus impact the people we hire or
promote to a higher function. This has a big impact on the glass ceiling (Brescoll, 2016). The fact that this stereotype affects the chance to be hired or promoted as a woman can also influence the performance and interest of women in these functions (Good, Aronson & Harder, 2008). This is mainly a problem in the scientific and technical sectors, but also in the business world. However, it is also a problem in the healthcare and education sector, where men are in the minority (Meadus, 2000; Mills, Martino & Lingard, 2004). The stereotype can influence men and women from a young age, because parents (unconsciously) project their expectations of men and women on their children, which can cause children to rate their abilities lower than they are (Jacobs & Eccles, 1992). The effect of this stereotype also has big implications for men. Because they are not expected to show sad feelings, they will often try and suppress those feelings. This can lead to mental health problems (Pennebaker, 1989).
The fact that our perception of the different genders is biased through our expectations can have big implications. If this stereotype also influences our short-term memories of people, then this means that there are more moments where this stereotype affects the choices we make, such as whom we hire or promote. Besides, if there are more moments where this stereotype influences our thoughts, then this stereotype will also affect the performance and interest of women. Of course, this is not the only thing that influences the existence of the glass ceiling, the pay gap, and mental health, but it is a significant factor.
This paper’s research did not show that expectations, based on the emotion-gender stereotype, caused short-term memory illusions. However, it can be concluded that the emotion-gender stereotype affects the perception. This means that our expectations of, in this case, the different genders do influence how we see people. The stronger our
expectations are, the higher the chances are that we will look for evidence for our
expectations and ignore the things that are not in line with our expectations. This is how the influence of our expectations on our perception can cause and even strengthen these
expectations (Otten, Seth & Pinto, 2017). The fact that stereotypes can bias our perception is an important finding, because this means that people will have to be very conscious of them not to have the stereotype influence the choices that are made.
For further research, it would be interesting to look at other stereotypes and their influence on the perception and/or the short-term memory or look at different stereotypical male and female emotions to see if those emotions affect short-term memory or perception.
Stereotypes can significantly influence our society, so it is crucial that people are aware of them. From this research, it could not be concluded that the emotion-gender stereotype influences the short-term memory, but it does influence our perception. More research has to be conducted to make sure that the stereotype does not cause short-term memory illusions.
References
Bernstein, D. M., & Loftus, E. F. (2009). How to Tell If a Particular Memory Is True or False.
Perspectives on Psychological Science, 4(4), 370–374. https://doi.org/10.1111/j.1745- 6924.2009.01140.x
Bhati, A. Does Implicit Color Bias Reduce Giving? Learnings from Fundraising Survey Using Implicit Association Test (IAT). Voluntas (2020). https://doi.org/10.1007/s11266-020-00277-8
Birnbaum, D., Nosanchuk, T. A., & Croll, W. L. (1980). Children’s stereotypes about sex differences in emotionality. Sex Roles, 6(3), 435–443.
https://doi.org/10.1007/bf00287363
Brescoll, V. L. (2016). Leading with their hearts? How gender stereotypes of emotion lead to biased evaluations of female leaders. The Leadership Quarterly, 27(3), 415–428. https://doi.org/10.1016/j.leaqua.2016.02.005
Carpenter, T., Pogacar, R., Pullig, C., Kouril, M., Aguilar, S., LaBouff, J. P., Isenberg., N., & Chakroff, A. (in press). Survey-software Implicit Association Tests: A methodological and empirical analysis. Behavior Research Methods.
Condry, J., & Condry, S. (1976). Sex Differences: A Study of the Eye of the Beholder. Child Development, 47(3), 812. https://doi.org/10.2307/1128199
Fabes, R. A., & Martin, C. L. (1991). Gender and Age Stereotypes of Emotionality. Personality and Social Psychology Bulletin, 17(5), 532–540.
https://doi.org/10.1177/0146167291175008
Flegal, K. E., Atkins, A. S., & Reuter-Lorenz, P. A. (2010). False memories seconds later: The rapid and compelling onset of illusory recognition. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 36(5), 1331–1338.
https://doi.org/10.1037/a0019903
Good, C., Aronson, J., & Harder, J. A. (2008). Problems in the pipeline: Stereotype threat and women’s achievement in high-level math courses. Journal of Applied Developmental
Psychology, 29(1), 17–28. https://doi.org/10.1016/j.appdev.2007.10.004
Greenwald, A. G., Nosek, B. A., & Banaji, M. R. (2003). Understanding and using the Implicit Association Test: I. An improved scoring algorithm. Journal of Personality and Social
Psychology, 85, 197–216. https://doi.org/10.1037/0022-3514.85.2.197
Heilman, M. E., & Okimoto, T. G. (2007). Why are women penalized for success at male tasks?: The implied communality deficit. Journal of Applied Psychology, 92(1), 81–92. https://doi.org/10.1037/0021-9010.92.1.81
expression: Are angry women more like men? Journal of Vision, 9(12), 19. https://doi.org/10.1167/9.12.19
Jacobs, J. E., & Eccles, J. S. (1992). The impact of mothers’ gender-role stereotypic beliefs on mothers’ and children’s ability perceptions. Journal of Personality and Social
Psychology, 63(6), 932–944. https://doi.org/10.1037/0022-3514.63.6.932 JASP Team (2020). JASP (Version 0.14)[Computer software]. Retrieved from
https://jasp-stats.org
Jeffreys, H. (1961). Small Corrections in the Theory of Surface Waves. Geophysical Journal
International, 6(1), 115–117. https://doi.org/10.1111/j.1365-246x.1961.tb02965.x Kelly, J. R., & Hutson-Comeaux, S. L. (1999). Gender-emotion stereotypes are context
specific. Sex Roles, 40(1/2), 107–120. https://doi.org/10.1023/a:1018834501996 Langner, O., Dotsch, R., Bijlstra, G., Wigboldus, D.H.J., Hawk, S.T., & van Knippenberg, A.
(2010). Presentation and validation of the Radboud Faces Database. Cognition &
Emotion, 24(8), 1377—1388. https://doi.org/10.1080/02699930903485076 Lenton, A. P., Blair, I. V., & Hastie, R. (2001). Illusions of Gender: Stereotypes Evoke False
Memories. Journal of Experimental Social Psychology, 37(1), 3–14. https://doi.org/10.1006/jesp.2000.1426
Lewis, K. M. (2000). When leaders display emotion: how followers respond to negative emotional expression of male and female leaders. Journal of Organizational Behavior,
21(2), 221–234. https://doi.org/10.1002/(SICI)1099-1379(200003)21:2<221::AID-JOB36>3.0.CO;2-0
Loftus, E. F., & Pickrell, J. E. (1995). The Formation of False Memories. Psychiatric Annals,
25(12), 720–725. https://doi.org/10.3928/0048-5713-19951201-07
Meadus, R. J. (2000). Men in Nursing: Barriers to Recruitment. Nursing Forum, 35(3), 5–12. https://doi.org/10.1111/j.1744-6198.2000.tb00998.x
Mills, M., Martino, W., & Lingard, B. (2004). Attracting, recruiting and retaining male teachers: policy issues in the male teacher debate. British Journal of Sociology of
Education, 25(3), 355–369. https://doi.org/10.1080/0142569042000216990
Neurotask. (z.d.). NeuroTask Online Testing. Retrieved in 2020, from https://neurotask.com Otten, M., Seth, A. K., & Pinto, Y. (2017). A social Bayesian brain: How social knowledge can
shape visual perception. Brain and Cognition, 112, 69–77. https://doi.org/10.1016/j.bandc.2016.05.002
Age, Gender, and Ambiguity on Emotion Perception. The Journal of Social Psychology,
154(4), 323–338. https://doi.org/10.1080/00224545.2014.901287
Pennebaker, J. W. (1989). Confession, Inhibition, and Disease. Advances in Experimental
Social Psychology, 211–244. https://doi.org/10.1016/s0065-2601(08)60309-3 Pinto, Y., Seth, A. & Otten, M. (in prep). Seeing Ɔ, remembering C: Illusions in short-term
memory
Plant, E. A., Hyde, J. S., Keltner, D., & Devine, P. G. (2000). The Gender Stereotyping of Emotions. Psychology of Women Quarterly, 24(1), 81–92.
https://doi.org/10.1111/j.1471-6402.2000.tb01024.x
[Qualtrics.com]. (z.d.) Retrieved in 2020, from https://qualtrics.com
Rouder, J. N., Speckman, P. L., Sun, D., Morey, R. D., & Iverson, G. (2009). Bayesian t tests for accepting and rejecting the null hypothesis. Psychonomic Bulletin & Review, 16(2), 225–237. https://doi.org/10.3758/pbr.16.2.225
Rudman, L. A., Greenwald, A. G., Mellott, D. S., & Schwartz, J. L. K. (1999). Measuring the Automatic Components of Prejudice: Flexibility and Generality of the Implicit Association Test. Social Cognition, 17(4), 437–465.
https://doi.org/10.1521/soco.1999.17.4.437
Schacter, D. L., Guerin, S. A., & St. Jacques, P. L. (2011). Memory distortion: an adaptive perspective. Trends in Cognitive Sciences, 15(10), 467–474.
https://doi.org/10.1016/j.tics.2011.08.004
Wagenmakers, E.-J., Wetzels, R., Borsboom, D., & van der Maas, H. L. J. (2011). Why
psychologists must change the way they analyze their data: The case of psi: Comment on Bem (2011). Journal of Personality and Social Psychology, 100(3), 426–432.
