« Go back to issue 44(5) summary

False memories in social anxiety disorder

Priscila de Camargo Palma1, Carmem Beatriz Neufeld1, Priscila Goergen Brust-Renck2, Carolina Prates Ferreira Rossetto1, José Alexandre de Souza Crippa1

1 University of São Paulo (USP), Ribeirão Preto, SP, Brazil.

2 Hospital of Clinics of Porto Alegre and University Center Ritter dos Reis, Porto Alegre, RS, Brazil.

Institution where the study was conducted: School of Philosophy, Sciences and Language of Ribeirão Preto, University of São Paulo, Ribeirão Preto, SP, Brazil.

Received: 04/13/2017– Accepted: 06/10/2017

DOI: 10.1590/0101-60830000000133

Address for correspondence: Carmem Beatriz Neufeld. Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto, Universidade de São Paulo. Av. Bandeirantes, 3900 – 14040-901 – Ribeirão Preto, SP, Brazil. Telephones: (+55 16) 3315-3724 / (+ 55 16) 3315-0197. Fax: (+55 16) 3315-4435. E-mail: cbneufeld@usp.br


Background: False memories are memories of events that never occurred or that occurred, but not exactly as we recall. Events with emotional content are subject to false memories production similar to neutral events. However, individual differences, such as the level of maladjustment and emotional instability characteristics of Social Anxiety Disorder (SAD), may interfere in the production of false memories. Objectives: This study aimed to assess the effect of emotion in memory performance for an event witnessed by participants with and without SAD. Methods: Participants were 61 young adults with SAD and 76 without any symptoms of SAD who were randomly assigned to watch a story with or without emotional arousal. Participants answered a subjective scale of emotion about the story and a recognition memory test. Results: Participants with SAD recovered more true memories and more false memories for the non-emotional version compared to the emotional version of the story. Overall, participants with SAD produced fewer false memories compared to those without SAD. Discussion: This finding suggests that social anxiety may have a significant impact on emotional memory accuracy, which may assist in the development and improvement of techniques for therapeutic intervention.

Palma PC et al. / Arch Clin Psychiatry. 2017;44(5):113-6

Keywords: False memories, emotional arousal, anxiety, social anxiety.


False memories, also referred to as distortions or illusions of memory, are memories of events that actually did not occur exactly as we remember1. This fact of recalling events that never occurred motivated several studies in the search for a greater understanding of these processes. It was observed that the level of emotional instability and mismatch significantly interferes in a greater number of false memories2,3, which are present in anxiety disorders.

Among the anxiety disorders, Social Anxiety Disorder (SAD) stands out, consisting of a marked and persistent fear of social or performance situations in which the individual might feel ashamed, being afraid of taking action where he/she may be humiliated and embarrassed. The social exposure or performance provokes a response of anxiety interfering significantly in the individual’s life4,5. It is estimated that the 12-month prevalence of SAD is 6.8% for the adult population6, one of the most common psychiatric disorders with a lifetime prevalence of 12%7.

It is known that in anxiogenic situations, perceptions of high probability of failure in performance are activated, which trigger cognitive, behavioral, and somatic symptoms causing a shift of attention from the environment to the individuals themselves, who remain in constant monitoring of their behavior, feelings, and physical sensations of anxiety8. This process is known as selective attention bias; anxious individuals initially direct their attention to a threat (e.g., emotional arousal with negative content), but then tend to shift attention in order to relieve sensations of anxiety, resulting in memory selectivity9,10.

Moreover, individuals with SAD tend to doubt their memories, interpreting ambiguous memories of social events in a less positive (or more negative) way compared to those without SAD11. One of the reasons that may compromise reliability of an individual’s memory is the belief that other people have a better memory12. One study found that less confident people (as in the case of individuals with SAD) accept the way the situation is recalled by another person (i.e., they acquiesce)2, which can lead to increased vulnerability to mnemonic distortions13.

SAD seems to be directly related to the fear of being rejected in social situations14. Individuals with a chronic level of maladjustment and emotional instability, notably insecure and inadequate, seem to be more easily swayed and susceptible to mnemonic distortions, such as false memories15,16. Recent studies have shown that individuals with psychosocial maladjustment produce more false memories3. Similarly, a desire for social interaction has been identified as a predictor of false memory for lists of semantically related words17. Despite these data, the literature still needs more research in clinical populations of individuals with anxiety disorders in general and with SAD, in particular.

Thus, the present study aims to investigate the performance of memory for events with and without emotional arousal, in a clinical sample of individuals with SAD. To do so, we used visual and audio material18,19, as these conditions are highly more likely to stimulate emotional arousal in this population8. The main hypothesis to be tested presumes that individuals with SAD are more prone to false memories compared to healthy subjects. Furthermore, it is known that the levels of false memories seem to be influenced by the presence of emotion, which will be assessed systematically in this study20,21.



The present study involved a 2 x 2 x 3 x 3 full factorial, mixed design, with repeated measures for the final two variables. The first variable was clinical characteristic of the population, in which some participants presented SAD and others had no clinical characteristic of anxiety. The second between subjects variable was the emotional arousal of the story, in which half the participants were exposed to a story with emotional arousal and the other half to a story without emotional arousal. Participants with and without SAD were randomly assigned to watch either version of the story. Both versions of the story were divided into three phases (phase 2 is where the versions of the story differed in terms of arousal) and into three types on the memory test (target, related distractor, and unrelated distractor). The dependent variables were the assessment target of the emotionality of the story (i.e., arousal), which was measured from the subjective scale of emotion, and memory performance for the story’s information in a recognition test.


Participants were 164 college students from Ribeirão Preto, São Paulo, Brazil who were diagnosed as part of a larger study on the prevalence of SAD22. All 2,319 students were contacted and invited to complete memory portion of the study.

Inclusion criteria for SAD sample were obtaining scores greater than or equal to 19 points in the Social Phobia Inventory (SPIN)22,23 and scores greater than or equal to 20 points in the Beck Anxiety Inventory (BAI)24, which indicate the presence of symptoms compatible with the diagnosis of SAD. The probable SAD individuals were evaluated by means of a version of the anxiety module of the Structured Clinical Interview for DSM-IV – Clinical Version (SCID-CV) translated and adapted to Portuguese25, for diagnostic confirmation. Furthermore, additional inclusion criteria for both SAD and non-SAD samples included no significant symptoms of depression based on scores of 6 or less and 10 or less in the Patient Health Questionnaire (PHQ-9)26 and the Beck Depression Inventory (BDI)24, respectively; good quality of life and well-being based on scores equal to or lower than 7 points on the Self-Reported Questionnaire (SRQ-20)27.

The final sample included 137 participants, 61 participants with SAD and 76 participants without symptoms of SAD. Among the participants with SAD, 98% did not use medication, 12% were in psychotherapy, no participant received a diagnosis of social phobia, 8% had a history of psychiatric disorder, and 18%, a history of psychiatric disorder in the family. Among the participants without SAD, 20% reported they had had some psychiatric symptom, however no one presented social phobia or was on medication, and 22% had a family history of psychiatric disorder. In addition, 6% of participants with SAD and 4% without SAD indicated that they were smokers or used drugs.

Participants were aged 17-34 years (M = 22.33; SD = 3.78), and 64% were female. Among the participants with SAD, 34 watched the emotional side of the story (mean age = 22.53, SD = 3.69), being 65% female. The number of participants with SAD who watched the non-emotional side of the story was 27 (mean age = 21.70, SD = 3.97), being 67% female. Among the participants without SAD, 44 watched the emotional version of the story (mean age = 23.32, SD = 4.12), and 64% were female. The number of participants without SAD who watched the non-emotional version was 32 (mean age = 21.28, SD = 2.92), being 59% female.


The memory of the participants was assessed through a slideshow procedure19, which was translated and adapted to Brazilian reality and to the investigation of false memories18, with methodological improvements28 for the memory test. The story consisted of 11 slides followed by a narrative divided into three phases. In phase 1 (slides 1-4), a mother and her son are on the way to the hospital where the father works; in phase 2 (slides 5-8), boy suffers an accident and goes to the hospital (arousal version) or the boy sees a crashed car and hospital procedures (non-arousal version); in phase 3 (slides 9-11), the mother goes away and calls home. Both versions had negative emotional content controlled for experimental purposes.

The emotional arousal of the story was assessed by a subjective 5-point Likert scale, ranging from “not emotional at all” (0) to “extremely emotional” (4).

Memory was tested by a self-administered memory recognition test composed of 84 yes-no items28. Participants were instructed to mark “yes” whenever the phrase corresponded to what was presented in the target material and “no” to all options that were not seen in the target material, even if it were information that could not be inferred about the event. The items of the test belonged to three categories: (i) target (equivalent to the measure of true memories); (ii) related distractor (equivalent to the measure of false memories); and (iii) unrelated distractor (equivalent to unrelated answers from error or “guess”). The items were distributed proportionally in relation to the three phases of the original material.


The study was approved by the Research Ethics Committee on Human Rights of the College of Philosophy, Sciences, and Letters of Ribeirão Preto – São Paulo University (USP) (N. CEP-FFCLRP nº 534/2010). All participants signed an informed consent form, prepared according to the ethical guidelines for research with human beings.

The memory material was presented with the use of multimedia equipment, and data collection was performed in groups in a silent room. Participants were advised to pay attention to the story, and to avoid comments throughout the procedure. Immediately after, subjects answered the subjective scale of emotion. Participants were instructed to be as honest as possible and to select the option that represents what they have really felt in terms of the arousal attributed to the emotion stimulated by the story. Before answering the memory recognition test, participants participated in a buffer activity. The buffer activity aimed to prevent the participant from retaining the online information.

Data were analyzed using SPSS Statistics, version 20. To analyze the data obtained from these instruments, we used Analysis of Variance (ANOVA) with Bonferroni correction for post hoc analysis. All statistical treatments used a α < 0.05 for hypothesis testing.

Results and discussion

An univariate ANOVA was conducted to test differences of emotional arousal between the versions of the story and presence of SAD. Results indicated a main effect of the version [F(1.133) = 36.31,
p < 0.001], demonstrating that the emotional version stimulates more arousal (M = 1.81; SD = 1.14) than the non-emotional version
(M = 0.80, SD = 0.72), which is consistent with previous studies18,28,29. In addition, we observed an effect of the presence of SAD [F(1.133) = 5.52, p < 0.05], which was responsible for a more intense evaluation of the effect of the stories’ emotional arousal (M = 1.58, SD = 1.06) in comparison to the participants without SAD (M = 1.22, SD = 1.11), which is consistent with typical aspects of this disorder, in which individuals often become hypervigilant regarding their physiological alterations and their bodily sensations, which seems to interfere in how participants with SAD rated the material30.

One hypothesis for this finding is that individuals with SAD believe that their appearance will accurately reflect all their physiological reactions of anxiety, which can be subject to scrutiny11, causing these individuals to utilize a protective measure, ignoring the physiological reactions experienced or by camouflaging them. Furthermore, one presumes that individuals with SAD get used to living with anxiety, so they somehow adapt to the physical reactions caused in situations that generate anxiety. As a result, subjects with SAD may have had difficulty in recognizing the emotions generated by the narrative.

Overall effects of memory performance were analyzed by means of an ANOVA with repeated measures for the presence of SAD and the version of the story for each item type. The rates of non-mnemonic responses (or guesses) were very low and there was no difference between the groups (ps > 0.001), suggesting that participants were paying attention to the task.

Regarding true memory, results showed higher rates for participants who watched the non-emotional version (M = 0.75, SD = 0.10) when compared to those who watched the emotional version of the story (M = 0.70, SD = 0.13) [F(1.133) = 6.10, p < 0,05]. Result were qualified by an interaction trend between presence of SAD and version of the story [F(1,133) = 2.97, p = 0,09]. Participants with SAD recovered more true memories for the non-emotional version (M = 0.76, SD = 0.09) than for the emotional version of the story (M = 0.68, SD = 0.13, p < 0,05). Given the typical processing of people with high levels of anxiety, it is possible to hypothesize that anticipatory anxiety prevailed over attentional processes30, causing individuals with SAD to remain more vigilant about the information of the video. This may have led to the lower rate of true memories for the emotional version.

With respect to false memories, there was a trend towards interaction between the presence of SAD and version of the story [F(1.133) = 3.76, p = 0.06], demonstrating increased production of false memories between the participants with SAD who watched the non-emotional version (M = 0.39, SD = 0.13) in comparison to those who watched the emotional version of the story (M = 0.31, SD = 0.15; p < 0.05). Among the participants who watched the non-emotional version, there was a trend toward greater production of false memories in participants with SAD in comparison to those without SAD (M = 0.32, SD = 0.12; p = 0.08).

These results suggest that memory from participants with SAD is affected by emotional nature of the story, which, in addition to their poor social skills (responsible for high levels of maladjustment and emotional instability) and attentional biases, made them vulnerable to selective memory performance. Overall, individuals with SAD exhibit impaired activation in the associative, interpretative, and attentional processes potentially leading to an attentional bias31,32, especially for stimuli perceived as dangerous or threatening33,34.

Anxiety disorders result in distortions in information processing, leading to perceptual sensitivity and memory bias34-36. This result seems to be in line with the findings of the study that proposes that individuals with SAD accept more outside influences when the situation is devoid of arousal, believing that the source of the material is more reliable than their own memory, and they reject more information of the memory test in general (both true and false)2.

Table 1. Mean of recognition (and standard deviation) by version of the story, type of item, and phase for participants with and without social anxiety disorder

Table 1. Mean of recognition (and standard deviation) by version of the story, type of item, and phase for participants with and without social anxiety disorder

An analysis of the results between the phases of the story was performed to isolate the effect of the arousal in the second phase of the story when arousal information was presented only in the emotional version. An ANOVA with repeated measures was performed for the presence of SAD and version of the story for each type of item (Table 1).

There was a main effect of phase for true memory [F(2.132) = 91.53, p < 0.001], indicating an increase in true memories in phase 2 in comparison to phase 3 (p < 0.001). This result was qualified by an interaction trend between the presence of SAD, version of the story, and phase [F(2.132) = 2.47, p = 0.09], in which we observed an increase in true memory in phase 2 compared to phase 3 under all conditions
(p < 0.05 for all cases). The effect of emotion from phase 2 was also observed in an interaction with presence of SAD, given that participants without SAD recovered more accurate information than participants with SAD (p < 0.05). Among participants with SAD, there was an increase in true memories for phase 2 for the non-emotional version of the story compared to the emotional version (p < 0.05).

With respect to false memories, there was a main effect of phase [F(2,132) = 116,17, p < 0.001]. The results showed that fewer false memories are produced in phase 2 than in phases 1 and 3
(ps < 0.001) and in phase 1 than in phase 3 (p < 0.05). In contrast, we observed an effect of phase on the production of non-mnemonic responses [F(2.132) = 3.69, p < 0.05], with lower indices for phase 2 than phase 1, although all indices were low in general. In this sense, the vulnerability of memory does not seem to be associated with emotional arousal for individuals with SAD. Therefore, the emotional arousal seems to be protective of memory in subjects with SAD, in line with the idea that anxious people effectively recall information related to risk34.

Finally, we can conclude that individuals with SAD present mnemonic changes. The results of this study help to further understand the attentional and mnemonic processes in individuals with SAD, and may assist in the development and improvement of techniques for therapeutic intervention. However, longitudinal studies, assessing memory performance before and after different treatment modalities (CBT, pharmacotherapy) and studies comparing SAD individuals to individuals with other anxiety disorders are deemed necessary and appropriate.


Preparation of this manuscript was supported in part by the São Paulo Research Foundation (Fapesp), Brazil, under Award Number 2010/11732-4 to the second author; a Masters Research fellowship by the Coordination for the Improvement of Higher Education Personnel (Capes), Brazil, to the first author; and a technical training fellowship from the São Paulo Research Foundation (Fapesp), Brazil, under Award Number 2011/08946-5 to the fourth author.


  1. Brainerd CJ, Reyna VF. The science of false memory. New York: Oxford University Press, 2005.
  2. Wright DB, London K, Waechter M. Social anxiety moderates memory conformity in adolescents. Appl Cogn Psychol. 2010;24(7):1034-45.
  3. Morgan J. Autobiographical memory biases in social anxiety. Clin Psychol Rev. 2010;30(3):288-97.
  4. Knappe S, Beesdo-Baum K, Fehm L, Stein MB, Lieb R, Wittchen HU. Social fear and social phobia types among community youth: differential clinical features and vulnerability factors. J Psychiatr Res. 2011;45(1):111-20.
  5. Schneier FR, Heimberg RG, Liebowitz MR, Blanco C, Gorenstein LA. Social anxiety and functional impairment in patients seeking surgical evaluation for hyperhidrosis. Compr Psychiatry. 2012;53(8):1181-6.
  6. Kessler RC, Chiu WT, Demler O, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62(6):617-27.
  7. Schneier FR. Social anxiety disorder. N Engl J Med. 2006;355(10):
  8. Makkar SR, Grisham JR. Social anxiety and the effects of negative self-imagery on emotion, cognition, and post-event processing. Behaviour research and therapy. Behav Res Ther. 2011;49(10):654-64.
  9. Liang CW, Hsu WY, Hung FC, Wang WT, Lin CH. Absence of a positive bias in social anxiety: The application of a directed forgetting paradigm. J Behav Ther Exp Psychiatry. 2011;42(2):204-10.
  10. Moscovitch DA, Suvak MK, Hofmann SG. Emotional response patterns during social threat in individuals with generalized social anxiety disorder and non-anxious controls. J Anxiety Disord. 2010;24(7):785-91.
  11. Hertel PT, Brozovich F, Joormann J, Gotlib IH. Biases in interpretation and memory in generalized social phobia. J Abnorm Psychol. 2008;117(2):278-88.
  12. Horry R, Palmer MA, Sexton ML, Brewer N. Memory conformity for confidently recognized items: The power of social influence on memory reports. J Exp Soc Psychol. 2012;48(3):783-6.
  13. Bernstein DM, Loftus EF. How to tell if a particular memory is true or false. Perspect Psychol Sci. 2009;4(4):370-4.
  14. Bögels SM, Alden L, Beidel DC, Clark LA, Pine DS, Stein MB, et al. Social anxiety disorder: questions and answers for the DSM‐V. Depress Anxiety. 2010;27(2):168-89.
  15. Kaplan RL, Van Damme I, Levine LJ, Loftus EF. Emotion and false memory. Emot Rev. 2016;8(1):8-13.
  16. Roediger III HL, McDermott KB. Tricks of memory. Curr Dir Psychol Sci. 2000;9(4):123-7.
  17. Neufeld CB, Brust-Renck PG, da Rocha AF, Sossella M, da Rosa FH. Falsas memorias e diferencas individuais: um estudo sobre fatores de personalidade e qualidade da memoria. Psicol Reflex Crit. 2013;26(2):319-27.
  18. Neufeld CB, Brust PG, Stein LM. Adaptação de um método de investigação do impacto da emoção na memória. PsicoUSF. 2008;13(1):21-9.
  19. Cahill L, Prins B, Weber M, McGaugh JL. Beta-adrenergic activation and memory for emotional events. Nature. 1994;371(6499):702-4.
  20. Kensinger EA. Remembering the details: Effects of emotion. Emot Rev. 2009;1(2):99-113.
  21. Brainerd CJ, Holliday RE, Reyna VF, Yang Y, Toglia MP. Developmental reversals in false memory: effects of emotional valence and arousal. J Exp Child Psychol. 2010;107(2):137-54.
  22. Osório FL, Crippa JAS, Zuardi AW, Graeff FG, Busatto G, Pinho M, et al. Inventário de Fobia Social (SPIN): Validação para o Brasil. In: XXII Congresso Brasileiro de Psiquiatria; 2004, Salvador-BA. Pesquisas Originais, Ensaios Clínicos e Revisão Sistemática.
  23. Connor KM, Kobak KA, Churchill LE, Katzelnick D, Davidson JR. Mini‐SPIN: A brief screening assessment for generalized social anxiety disorder. Depress Anxiety. 2001;14(2):137-40.
  24. Cunha JA. Manual da versão em português das escalas. São Paulo: Casa do Psicólogo, 2001.
  25. Del-Ben CM, Vilela JA, de S Crippa JA, Hallak JE, Labate CM, Zuardi AW. Confiabilidade de “Entrevista Clínica Estruturada para o DSM-IV–Versão Clínica” traduzida para o português. Rev Bras Psiquiatr. 2001;23(3):156-9.
  26. Osório FL, Mendes AV, Crippa JA, Loureiro SR. Study of the Discriminative Validity of the PHQ‐9 and PHQ‐2 in a Sample of Brazilian Women in the Context of Primary Health Care. Perspect Psychiatr Care. 2009;45(3):216-27.
  27. Mari JJ, Williams P. A validity study of a psychiatric screening questionnaire (SRQ-20) in primary care in the city of Sao Paulo. Br J Psychiatry. 1986;148:23-6.
  28. Barbosa ME, Brust-Renck PG, Stein LM. O papel do alerta nas memórias verdadeiras e falsas para informações centrais e periféricas. Psicol Reflex Crit. 2014;27(1):100-9.
  29. Gasbarri A, Arnone B, Pompili A, Marchetti A, Pacitti F, Calil SS, et al. Sex-related lateralized effect of emotional content on declarative memory: An event related potential study. Behav Brain Res. 2006;168(2):177-84.
  30. Clark DA, Beck AT. The anxiety and worry workbook: the cognitive behavioral solution. Guilford Press; 2011.
  31. Indovina I, Robbins TW, Núñez-Elizalde AO, Dunn BD, Bishop SJ. Fear-conditioning mechanisms associated with trait vulnerability to anxiety in humans. Neuron. 2011;69(3):563-71.
  32. Shin LM, Liberzon I. The neurocircuitry of fear, stress, and anxiety disorders. Neuropsychopharmacology. 2010;35(1):169-91.
  33. Papageorgiou C, Wells A. Process and meta‐cognitive dimensions of depressive and anxious thoughts and relationships with emotional intensity. Clin Psychol Psychother. 1999;6(2):156-62.
  34. Jansson B, Najström M. Is preattentive bias predictive of autonomic reactivity in response to a stressor?. J Anxiety Disord. 2009;23(3):374-80.
  35. Beck AT, Clark DA. An information processing model of anxiety: automatic and strategic processes. Behav Res Ther. 1997;35(1):49-58.
  36. Bishop SJ. Neurocognitive mechanisms of anxiety: an integrative account. Trends Cogn Sci. 2007;11(7):307-16.