Update of the Brazilian norms for the International Affective Picture System (IAPS)

Abstracts

This study includes the standardization of 240 new stimuli in the Brazilian norms for the International Affective Picture System (IAPS), an instrument of affective images widely used in research, and confirms the validity of the categories of "pleasant," "neutral" and "unpleasant" pictures. A comparison was performed of the scores of affective ratings between the male and female samples of university students (179 males, and 269 females), who assessed the affective content of photographs in terms of valence, arousal and dominance using the Self Assessment Manikin (SAM) scale. The methods were the same used in a study conducted for the American norms.

IAPS; emotion; sex differences; valence; arousal

Este trabalho expande a normatização brasileira do International Affective Picture System (IAPS), um instrumento de imagens afetivas, amplamente utilizado em pesquisas, com valores normativos para 240 novos estímulos, e confirma as categorias de imagens "agradáveis", "neutras" e "desagradáveis". Também são apresentados os resultados comparando os padrões de respostas de homens e mulheres. No total, participaram 448 estudantes universitários brasileiros (179 homens e 269 mulheres), que avaliaram as fotografias em termos de prazer, alerta e dominância, através da escala Self Assessment Manikin (SAM). O procedimento adotado foi o mesmo de outro estudo já desenvolvido para as normas norte-americanas.

IAPS; emoção; diferenças sexuais; prazer; alerta

BRIEF COMMUNICATION

Update of the Brazilian norms for the International Affective Picture System (IAPS)^*

Cristina Lasaitis^I; Rafaela Larsen Ribeiro^II; Marcelo Ventura Freire^III; Orlando Franscisco Amodeo Bueno^IV

^IMaster's Degree student, Graduate Program in Psychobiology, Universidade Federal de São Paulo & Escola Paulista de Medicina (UNIFESP-EPM), São Paulo, SP, Brazil.

^IIPhD. Associate professor, Department of Psychobiology, UNIFESP-EPM.

^IIIPost-PhD. Professor, School of Arts, Sciences and Humanities, Universidade de São Paulo (USP), São Paulo, SP, Brazil.

^IVPhD. Professor. Associate professor, Graduate Program in Psychobiology, UNIFESP-EPM.

Correspondence

ABSTRACT

This study includes the standardization of 240 new stimuli in the Brazilian norms for the International Affective Picture System (IAPS), an instrument of affective images widely used in research, and confirms the validity of the categories of pleasant, neutral and unpleasant pictures. A comparison was performed of the scores of affective ratings between the male and female samples of university students (179 males, and 269 females), who assessed the affective content of photographs in terms of valence, arousal and dominance employing the Self Assessment Manikin (SAM) scale. The methods were the same used in a study conducted for the American norms.

Keywords: IAPS, emotion, sex differences, valence, arousal.

Introduction

The International Affective Picture System (IAPS)¹ is an instrument of affective images representing different events of life and values broadcast by the media and present in the general culture of many countries (such as family, sports, sexuality, violence, health, etc). Since it is an efficient instrument to elicit different emotional states and because it permits a high level of control regarding the parameters of exposure and measurement in experimental contexts, the IAPS has been widely used in studies of cognition, affection, behavior, social attitudes, psychophysiology, among others, with the purpose of studying patients with neuropsychiatric disorders, as well as healthy populations.

Some studies have demonstrated that the main factors involved in the emotional experience are valence and arousal, which constitute an affective space representing the hedonic valence and the motivational activation.² Within this bidimensional space, the ratings of positive and negative valence represent the appetitive-aversive system, whereas the rating of arousal indicates the intensity of the motivational activation.³ The pictures included in the IAPS are distributed throughout this affective space of valence and arousal and they can be classified according to the psychophysiological states they elicit as follows: pleasant arousing, pleasant relaxing, neutral and unpleasant pictures.⁴ Curiously, the instrument does not include relaxing unpleasant pictures. A third dimension, dominance, represents the level of control of the emotional response (which ranges from submissive to dominant), and it has also been commonly used to assess IAPS stimuli.

For the subjective classification of IAPS stimuli, we used the Self Assessment Manikin (SAM),⁵ a self-report instrument containing three scales, one for each dimension studied (valence, arousal and dominance). Each dimension is portrayed by five pictures that symbolize different emotional levels, for instance: the valence dimension is represented by a sad/unhappy doll at one end and a happy/smiling doll at the other end, and its respective degrees are inserted between these limits. The subjects respond to the pictures by making an X on one of the five pictures in each scale or on the spaces between the pictures, resulting in nine possible answers. The lowest score for each dimension is 1, which indicates the lowest level of valence, arousal or dominance, whereas 9 is the highest score and it represents the highest level of the same dimensions. The SAM scales are easily understandable and can be employed to measure emotional responses to different stimuli involving different types of subjects.

In human beings, the emotional experience cannot be restricted to the biological determinism as if it were a product generated only by the motivational circuits, since it is also affected by other factors, including personal, social, cultural or contextual aspects. However, countless languages and different sociocultural contexts can stereotype emotions and corroborate the hypothesis of a universal biological determinism. A consequence of that fact is that, despite slight differences, the pattern of response to IAPS stimuli through the valence and arousal dimensions has proven to be very similar in the different populations studied, as demonstrated when we compare the scores of the American¹ and Spanish^6,7 norms with the Brazilian norm.^8,9 Once IAPS is an instrument in constant development, new sets of pictures have been included since it was first designed. The Brazilian standardization, which was first performed by Ribeiro et al.,^8,9 established the norms for 707 stimuli related to sets 1 to 12 of the IAPS.

The objectives of this study were: 1) to extend the Brazilian standardization of the IAPS to the 240 more recent stimuli that constitute sets 13 to 16 of the original study; 2) to confirm the validity of the categories of pictures; 3) to compare the different patterns of emotional response between Brazilian men and women.

Methods

Subjects

This study included 448 university students (179 men and 269 women), whose ages ranged from 18 to 62 years (mean 24.2±7,8), and were all native speakers of Portuguese. They were selected from different majors (medicine, biomedicine, psychology, accounting, business administration and physical education) at public and private universities in the city of São Paulo. The selection was performed by means of previous contacts with university professors who agreed with the use of classrooms and projectors so that the study could be carried out with groups of students. No information about the subjects' physical or mental status was collected; however, data from those volunteers who did not have corrected vision, did not report their sex or were negligent while filling out the scales were excluded from the analysis, which accounted for a total of 9.3% of the participants.

Material

Two hundred and forty IAPS pictures belonging to sets 13 to 16 (described by Lang et al.¹) were used.

Procedure

The procedures employed and the analyses of the results were the same of the previous study,^8,9 which, on its turn, was based in the original study.¹ In each experimental session, 30 randomly selected IAPS pictures from sets 13 to 16 were presented. A printed version of the SAM scales (paper and pen version) was handed to the subjects for the evaluation of the stimuli.⁵ It allowed the subjects to classify each picture in three scales of 9 points. Which are: valence (ranging from 1 for pleasant to 9 for unpleasant), arousal (ranging from 1 for calm/relaxed to 9 for aroused/excited) and dominance (ranging from 1 for submissive to 9 for dominant).

Each experimental session included a projection of slides containing instructions and examples of scores of the SAM scale for each dimension (valence, arousal and dominance). In addition to the 30 IAPS pictures classified in each session, three pictures were included as examples used by Lang. These pictures were previously displayed so that the subjects could evaluate them in a testing sheet.

Before each session, a preliminary projection of slides containing instructions and examples of classifications in the valence, arousal and dominance dimensions was performed. A preparation screen showed the number of the slide (from 1 to 30) during 5 seconds, after that period of time, a picture was displayed during 5 seconds and, during the next 15 seconds, none picture was shown so that the subjects could take note of their classifications on the test sheet. After that, the researcher announced the next slide outloud, and a new preparation screen was shown. Sessions were carried out in classrooms where there were 13 to 60 students. The classrooms had appropriate lighting so that the slides could be easily seen.

Data analysis

Means and standard deviations of the affective ratings of each picture were determined for all subjects and, separately, for men and women in each dimension. Since the pictures were randomly distributed among the eight sets studied, the scores for the pictures were rearranged according to the sets belonging to the original study during the statistical analysis. Student's t tests were performed to compare men and women's means for each picture. The significance level was 5%.

In addition to the classification of the pictures as pleasant, neutral and unpleasant according to the study by Ribeiro et al.,⁸ a cluster analysis was carried out to confirm if the adequacy of these categories is corroborated by the respondents' perceptions of the pictures.

Results

The results were consistent with the data obtained in the first study.^8,9 Women, when compared to men, provided lower scores of pleasure and higher scores of arousal for the whole set of pictures, but only in the dominance dimension there were significant differences, since women reported lower scores than men (Table 1). Another relevant result is related to the fact that there was no consensus between sexes regarding the pictures classified as pleasant (P - valence > 6), unpleasant (U - valence < 4) and neutral (N - valence between 4 and 6). We found that women classified fewer pictures as neutral and more pictures as pleasant and unpleasant in comparison to men (Table 2). Considering these classifications, there were no differences in the means of valence, arousal and dominance for neutral and pleasant pictures. However, for those pictures considered as unpleasant, there were significant differences in the three dimensions, as women reported lower scores of valence and dominance and higher scores of arousal.

In the cluster analysis, considering the male and female means of valence, arousal and dominance at the same time, initially we just detected as clearly different one more homogenous cluster and one large and more heterogeneous cluster with great internal variability, which is a characteristic of not well determined clusters, although they are identified as only one cluster (Figure 1). The ETA2 index of proportion of variation explained suggests that at least three clusters are necessary to explain at least 80% of the variability present in the data, and the PRE index of proportional reduction in variation suggests that there will not be a high marginal gain if more than four clusters are used (Figure 2). The means of the three clusters (Figure 3) suggest that the first cluster (which is the most evident; n = 79) can be identified as the one comprising the unpleasant pictures, whereas the second cluster (n= 97) contains the neutral pictures, and the third cluster (n = 64) comprises the pleasant pictures.

The normative scores of the Brazilian standardization of the pictures that constitute sets 13 to 16 of the IAPS, as well as the data obtained in the first study related to sets 1 to 12, are available at the website of Universidade Federal de São Paulo (UNIFESP) : www.unifesp.br/dpsicobio/adap/adapta.htm.

Discussion

The affective ratings were different for men and women regarding the pictures classified as unpleasant. Women provided higher scores of arousal and lower scores of valence and dominance to these pictures. Women, when compared to men, classified more pictures as pleasant and unpleasant, and fewer pictures as neutral, which might be due to a sharper perception of the emotional valence. Such female characteristic has an impact on the level of arousal, which is highly correlated with valence. In addition, this different pattern of response between sexes was also described in the American and Spanish norms for the IAPS,^1,6,7 and it is consistent with other findings in the literature.¹⁰ These data corroborate the finding that the emotional memory, as well as the subjective experience of emotions, is more stressed in women, which might explain the higher incidence of anxiety-depression disorders in the female population.

Our analysis also found that IAPS stimuli can be grouped in three clusters, confirming the validity of the three categories of pictures described (since the previous study^8,9) as "pleasant", "neutral" and "unpleasant", and demonstrating that these categories of pleasantness are not exclusively dependent on the emotional valence, but are influenced by other relevant factors in the experience of emotions, such as the level of arousal and dominance.

The IAPS is one of the most widely used tools in the world to elicit affective states currently available for research. However, it is barely known and scarcely used by Brazilian researchers. Combined with the SAM, IAPS is a very versatile tool for emotional assessment. Its simple application and easy comprehensibility make its use possible in a large variety of experimental subjects, and allows an efficient collection of data, based on a tridimensional analysis of emotions, presenting the possibility of exploring complex aspects of motivation and behavior, such as affective valence, emotional arousal and dominance. Hence, it is important to make the IAPS normative scores for the Brazilian population available to all interested researchers and professionals, thus studies in the field of cognition, emotion and neuropsychiatric disorders can be carried out.

In conclusion, the present study extends the Brazilian standardization of the IAPS, adding the normative scores of 240 new stimuli and confirming their experimental validity within preestablished emotional categories. The scores found have excellent stability regarding the first part of the Brazilian standardization and good consistency with the data obtained in the original standardization, which reinforces the reliability of the IAPS as a psychometric instrument.

Acknowledgements

We are thankful to FAPESP (Project no. 2006/05144-7) and to AFIP for financial support. We also thank the professors and students from the following universities: UNICID, UNIFESP, UniFMU, PUC, São Luís, São Marcos, Faculdades Paulistanas (FAPA) e Fundação Escola de Comércio Álvares Penteado (FECAP) for their assistance in data collection.

References

• 1. Lang PJ, Bradley MM, Cuthbert BN. International Affective Picture System (IAPS): Affective ratings of pictures and instruction manual. Technical Report A-6. Gainesville: University of Florida; 2005. http://www.unifesp.br/dpsicobio/adap/instructions.pdf Acessado dez 2008.
• 2. Mehrabian A, Russel JA. An approach to environmental psychology. Cambridge: MIT; 1974.
• 3. Bradley MM, Codispoti M, Cuthbert BN, Lang PJ. Emotion and motivation I: defensive and appetitive reactions in picture processing. Emotion. 2001;1(3):276-98.
• 4. Ribeiro RL, Teixeira-Silva F, Pompéia S, Bueno OFA. IAPS includes photographs that elicit low-arousal physiological responses in healthy volunteers. Physiol Behav. 2007;91(5):671-5.
• 5. Lang PJ. Behavioral treatment and behavioral assessment: computer applications. In: Sidowski JB, Johnson JH, Williams TA, eds. Technology in mental health care delivery systems. Norwood: Ablex; 1980. p. 119-37.
• 6. Moltó J, Montañés S, Poy R, Segarra P, Pastor MC, Torno MP, et al. Un nuevo método para el estudio experimental de las emociones: Int Affect Pict Syst (IAPS): Adaptación Española. Rev Psicol Gen Aplicada. 1999;52(1):55-87.
• 7. Vila S, Sánchez M, Ramírez I, Fernández MC, Cobos P, Rodríguez S, et al. El Sistema Internacional de Imágenes Afectivas (IAPS). Adaptación española. Segunda Parte. Rev Psicol Gen Aplicada. 1999;54(4):635-57.
• 8. Ribeiro RL, Pompéia S, Bueno OFA. Brazilian norms for the International Affective Picture System (IAPS): brief report. Rev Psiquiatr RS. 2004;26(2):190-4.
• 9. Ribeiro RL, Pompéia S, Bueno OFA. Comparison of Brazilian and North-American Norms for the International Affective Picture System (IAPS). Rev Bras Psiquiatr. 2005;27:208-15. http://www.scielo.br/pdf/rbp/v27n3/a09v27n3.pdf Acessado dez 2008.
• 10. Canli T, Desmond JE, Zhao Z, Gabriel JD. Sex differences in the neural basis of emotional memories. Proc Natl Acad Sci USA. 2002;99(16):10789-94.

*

Apoio financeiro: FAPESP (processo n° 2006/05144-7) e AFIP.

Publication Dates

• Publication in this collection
  17 Mar 2009
• Date of issue
  Dec 2008

History

• Received
  14 Apr 2008
• Accepted
  01 Sept 2008