Subjective affective ratings to photographic stimuli of the International Affective Picture System in a Brazilian elderly sample

Abstracts

INTRODUÇÃO: A literatura científica indica a possibilidade de a percepção da emoção e a formação da memória emocional serem discordantes entre jovens e idosos. A mesma é pobre ao explorar essa possibilidade. Neste estudo, relatamos os resultados obtidos em um experimento-piloto com uma amostra de idosos brasileiros, que classificaram subjetivamente, através da escala Self Assessment Manikin, imagens oriundas do International Affective Picture System. MÉTODO: Quarenta e oito idosos voluntários da Universidade Aberta da Terceira Idade, saudáveis clínica e cognitivamente, avaliaram o caráter alertante e a valência afetiva de 71 imagens do International Affective Picture System, aleatoriamente escolhidas. RESULTADOS: O grau de alerta reportado por idosos diante de um estímulo emocional é tanto maior quanto menor o prazer provocado por essa imagem-estímulo, resultando na existência de uma forte correlação negativa (r = 0,93) entre o grau de alerta e o estímulo desprazeroso. Em uma comparação do acima obtido com outro experimento normativo semelhante feito com jovens brasileiros e americanos, apontou-se para uma possível diferença cultural na forma de relatar subjetivamente um estímulo emocional. CONCLUSÕES: Os resultados obtidos com esta amostra estudada sugerem que pode existir uma diferença nos relatos afetivos entre os jovens e idosos, onde uma normatização do International Affective Picture System para uma amostra maior, representativa da população de idosos, seria útil para responder esta questão.

International Affective Picture System; Self Assessment Manikin; idosos; Brasil

en_v30n2a09

ORIGINAL ARTICLE

Subjective affective ratings to photographic stimuli of the International Affective Picture System in a Brazilian elderly sample

Weyler Galvão Porto^I; Paulo Bertolucci^II; Rafaela Larsen Ribeiro^III; Orlando Francisco Amodeo Bueno^IV

^IPhysician. MSc. in Gerontology. PhD in Psychobiology. Coordinator, Memory, Behavior and Mood Disorders Clinic, Department of Geriatrics, Hospital Estadual Mário Covas, Fundação do ABC, Santo André, SP, Brazil.

^IIPhD. Head, Department of Behavioral Neurology, Universidade Federal de São Paulo (UNIFESP), São Paulo, SP, Brazil.

^IIIPhD. Visiting associate professor, Department of Psychobiology, UNIFESP.

^IVProfessor. Associate professor, Department of Psychobiology, UNIFESP.

Correspondence

ABSTRACT

INTRODUCTION: The scientific literature points to a possible bias in the form perception and emotional memory are constructed when elderly and young individuals are compared. However, this possibility is underexplored. This paper presents the results obtained from a pilot study based on an elderly emotional subjective report after evaluation using the International Affective Picture System images and Self Assessment Manikin scale scores.

METHOD: Forty-eight clinically and cognitively capable elderly volunteer subjects from the Third Age Open University evaluated 71 randomly chosen images of the International Affective Picture System in terms of arousal and affective valence.

RESULTS: For the elderly, the greater the arousal, the smaller the pleasure resulting in a strong negative correlation (r = 0.93) observed between arousal and negative valence. A comparison with another similar normative experiment performed in young Brazilian and American individuals showed a possible cultural difference in subjective reports of emotional stimuli.

CONCLUSIONS: This investigation indicates that there may be a difference between elderly and young individuals when affective reports of arousal are studied. A normalization of the International Affective Picture System for the elderly in a larger sample, representative of the population, might be useful to address this issue.

Keywords: International Affective Picture System, Self Assessment Manikin, elderly, Brazil.

INTRODUCTION

Epidemiologic and demographic revolutions that culminated in human population aging introduced new paradigms in neurosciences. So far, most scientific work has only included youths and children. However, there is the hypothesis that not all results, material and methods obtained from studies on youths can be extrapolated directly to geriatrics and gerontology without previous adaptations to the needs of elderly individuals.^1-5 In general, science has a poor explanation for this issue, and the situation is even more complex when dealing with emotional memory of this demographic portion.^2,5

Aging in general is associated with a period of decline in human evolution. The elderly may have cognitive deficits as a result of physiological changes that are intrinsically related to the aging process; systemic diseases that compromise the central nervous system (CNS), in addition to primary diseases of the central, autonomous and peripheral system. However, there is a group of individuals that does not fit in any of the categories mentioned above and grow old successfully. These are considered healthy and totally inserted in their social contexts. Even so, there is a primary cognitive deficit associated with the aging process (CDAP). CDAP is characterized by a reduction in processing speed, decline in fluid intelligence and episodic memory deficit.^1,6 These symptoms can be responsible for cognitive error commonly presented by elderly individuals. Among them is improper recognition of semantically related words and figures of similar categories; reduced ability to discriminate new and old details in free memory tasks; confusions of stereotyped situations as part of another previously presented situation; and error in source attribution after a given piece of information.⁷

On the other hand, the effect of the alert level among CDAP characteristics is also questioned - and even its effect on declarative memory of elderly considered as healthy. Doubts about the emotional memory of elderly individuals are even more frequent. Following this line, some authors have showed that when the elderly are compared with the youth, a) they acquire control and regulation over their emotional well-being, b) emotions gain a central role in the cognitive process, and c) they have relative preservation of anatomical areas involved in the alert process. Apparently, therefore, not all is decline in cognition and emotion among the elderly, and this fact should be more properly explored in controlled studies on emotion and aging.^4,5,8

Emotions can be considered as measurable biological phenomena. In general, emotional phenomena can be expressed and then measured in three categories: 1) functional emotional behaviors presented as behavioral acts and sequences; 2) emotional language; and 3) physiological reactions.^9-12

Nowadays, the perfect method to replicate emotions in controlled studies, exactly as they occur in real life, is far from what may be dreamed as the gold standard. The methodologies causing fear and alertness in elderly individuals are limited by obvious ethical aspects. For this purpose, search for autobiographic memories can normally be used (conscious or induced by hypnosis), mood and/or emotion induced situations or recreations of phobic environments through photos, videos or words.^13,14 The International Affective Picture System (IAPS),¹⁵ the International Affective Digitized Sound System (IADS) and the Affective Lexicon of English Words (ANEW), among other methods, aim at recreating an emotional atmosphere by using emotional pictures, sounds and/or words. The emotional response aroused by these methods is measured through subjective affective reports, measured at an analogical scale known as Self Assessment Manikin (SAM), and/or may also be objectively measured by physiological and behavioral responses.

Among the measurement instruments presented above, the IAPS has specific characteristics when used by different countries, as has been shown in Spanish, Brazilian and American studies. In general, it is concluded that the same emotional stimulus can be perceived differently by distinct cultures.^16,17 In those studies there is an adaptation proposal for the IAPS for each population target or sample to be investigated in the sense of predicting and avoiding eventual errors of preconceptual conclusion. Factors such as age, diseases, schooling, among so many others, may interfere with evaluation of affective stimulus.

Therefore, this pilot study was developed with the aim of exploring the existence of a possible theoretical difference, as discussed above, in the evaluation of subjective affective ratings between Brazilian elderly and young individuals, using IAPS images as emotional stimuli.

METHOD

Population

The pilot project was developed by the Department of Psychobiology and by the Department of Neurology of the Behavioral Neurology Sector at Universidade Federal de São Paulo (UNIFESP) and voluntarily included students of Third Age Open University (TAOU). TAOU has 175 students over 60 years of age regularly enrolled; 83% are women, and 17% are men. Of this population, a sample of 48 students (27.43%) was chosen by simple random sampling - 42 women (87.5%) and six men (12.5%) - and were not statistically different as to age and schooling - 68.65±6.7 years (mean ± SD), 8 years of schooling. All the participants wore glasses with correction lenses.

They were all independent for basic (BDA) and instrumental (IBDA) daily life activities; 100% of the individuals had six points out of six (6/6) for BDA and eight out of eight (8/8) for IBDA.¹⁸ They had no clinical history of neoplasms, metabolic, infectious or psychiatric disorders, traumatic brain injury, loss of consciousness, abuse of illicit drugs or medications that could change cognition. Their results in Hachinski ischemic scale were < 4,¹⁹ and the score in the spontaneous clock drawing test was 12.7±2.6.^20,21

Material

Forty slides were initially selected, and later 71 were chosen from the IAPS¹⁵ by simple random process.

Procedure

In the first stage, the first 40 slides were chosen and divided into two subgroups of 20 images. Each subgroup was balanced as to affective valence and alert level so that there were no affective extremes between groups. The group of 48 elderly was also randomly divided into two subgroups. A subgroup of 20 images was presented to 24 elderly individuals (subgroup A), who were asked to watch the images for 5 seconds and classify them within 10 seconds according to the application manual produced for a sample of youths.^15,17 The other half was presented to the remaining 24 elderly individuals (subgroup B).

In a second stage, 71 extra slides were randomly chosen and presented in a balanced manner, in groups of 10 individuals/session, and were later analyzed according to the methodology in the application manual and in the norms for Brazilian youths adapted for elderly patients after the previous stage. Each of the images belonging to the IAPS was given an SMA score relative to the dimensions affective valence and alert level, determined by each participant in the study. Mean of these scores determines the subjective affective value of each image reported by the elderly. The values obtained from SAM were then compared with their corresponding values obtained from studies in young individuals.

At the beginning of each session, the individuals signed a consent form approved by the Ethics Committee of UNIFESP, in accordance with resolution CEP nº 438/01, June 11, 2001, and all questions were answered by the researcher.

Statistical analysis

The software Statistica (version 5, 1997) was used to statistically analyze all the material and data. It was arbitrarily defined that SAM scores for the purpose of calculations should have the following values: pleasure: 6.01-9; neutral: 3.01-6; displeasure: 1-3; alert: 6.01-9; relaxation: 1-3.

Measurable units evaluated in the statistical analysis were the means of means obtained in each group and between them. The groups did not have a normal statistical curve or homoscedasticity. Therefore, a non-parametric methodology for variance analysis - Kruskal-Wallis test, Mann-Whitney U test and Spearman's correlation coefficient (r_s) was used. Significance level was α₂ = 0.05.

RESULTS

The first stage was used to adapt the pen-and-pencil version of the IAPS¹⁵ to the need of elderly individuals. Subgroup A, which evaluated 20 images strictly following the guidance contained in the original American manuscript,¹⁵ was not able to properly recognize or describe images with emotional content (p < 0.05) when compared with subgroup B, which accurately recognized and described more than 75% of images with emotional content after the changes performed in the manual, such as size of projected image, time of image projection, room lighting, etc. These changes can be found in the adaptation of the Manual for IAPS Application in Brazilian Elderly, available at http://www.unifesp.br/dpsicobio/adap/adapta.htm.

In the second stage, after changes in the manual were performed, the same 48 elderly individuals were submitted to evaluation of the remaining 71 slides from IAPS according to SAM emotional classification. These SAM scores regarding alert level and affective valence are shown in Table 1.

Comparison of means relative to affective valence (Table 2) between the sample of Brazilian elderly and Brazilian and American youths had no statistically significant difference - Kruskal-Wallis test ==> H (2, N = 213) = 1.331; p = 0.5140. However, there was statistical difference in the comparison of alert level between groups - Kruskal-Wallis test ==> H (2, N = 213) = 11.02; p = 0.004] -, especially regarding the comparison of Brazilian youths and elderly studying at TAOU (Mann-Whitney ==> U = 1838; Z = -2.78; p = 0.0054) and Brazilian and American youths (Mann-Whitney ==> U = 1802; Z = -2.93; p = 0.0034). However, when the elderly at TAOU and American youths are compared, this difference is no longer statistically significant (Mann-Whitney ==> U = 2421; Z = 0.41; p = 0.685). These results can be seen in Table 3.

Table 4, in turn, shows the results of a strong Spearman's negative correlation (r_s) between alert level and affective valence in the three samples.

Figure 1 shows a scatter diagram (formed by the alert level and affective valence axes and named affective space) obtained by analysis of subjective affective ratings of elderly individuals at TAOU.

Finally, a frequency test (χ²) showed a statistically significant difference as to alert level for Brazilian elderly in relation to Brazilian youths (χ² p = 0.01). What is neutral for the elderly may not be for youths. The same is not true for affective valence (χ² p = 0.07) in both groups.

DISCUSSION

The material presented here is a result of a pilot project, characterized as a fragment of a longer study,²² whose basic objective is to explore how normal elderly individuals evaluate, through subjective affective ratings, a volume of IAPS images loaded with emotion. A comparison between these subjective ratings by elderly individuals and those performed by youths over the same period is also of great interest. At the time, a parallel study was developed including youths¹⁷ in our department.

As mentioned earlier, most studies conducted with IAPS images only include youths. In the literature relative to these studies, there is a diagram called affective space (Figure 1),^15-17 which is usually quite useful for the definition of affective rating pattern in different samples. The figure of affective space obtained by the elderly in this study tends to linearity focused on the poles, different from the results obtained by youths. The polarized figure seems to suggest that unpleasant images alert, whereas pleasant images relax the elderly faced with images with emotional content. They could also suggest that these same images may tend to cause "all-or-nothing" affective responses, emphasizing emotional extremes of affective valence and alert level to the detriment of neutral details. The strong negative correlation between displeasure level and alert level in this pilot study indicates a statistical association between these variables. The feasibility of these conclusions can be corroborated by similar studies in the literature, suggesting that differences in the nature of affective dimension can be obtained from variables such as age, psychopathy or brain lesion. They also suggest that alerting stimuli are rated as unpleasant for the elderly and for depressive patients. On the other hand, pleasant alerting images do not seem to be relevant to this category.^11,23

The results obtained in this work are in accordance with other studies showing that emotional rating gains privileged value in the elderly's everyday life.⁵ According to the literature, the elderly tend to emphasize emotional traits of a hypothetical situation to the detriment of details. This age group tends to proportionally remember more information with emotional traits than contextual and spatial details in tests of free memory recall when compared with youths. This characteristic is closely related with age group. The older the individuals, the higher number of emotion traits will be remembered. Therefore, it makes sense to have an "all-or-nothing" affective space diagram (Figure 1). A large amount of the 71 images considered here migrate from their original neutral state observed in youths to subjective affective extremes. The frequencies observed as to their alert level in the group of elderly individuals are statistically different from those expected for youths. The same is not true for the dimension affective valence.

Therefore, how can the elderly biologically benefit from this different form of perceiving emotion? Research indicates that youths seem to have more ability in regulating their emotions than a population of youths. Their strategy to change the impact of emotional response is to focus on a response before generating such emotion.²⁴ Regulation of emotional well-being can be well balanced between elderly adults and even better in some situations when compared with that of the young population.²⁵ The socioemotional selectivity theory (SST) supports that the elderly give more value to objectives with emotional meaning and invest their strengths more cognitively and behaviorally aiming at these targets.²⁶ Thus, emotion may direct cognitive processes and, by directing them, the memory formed could be influenced in its content.

Could the elderly be directed by evaluation of positive or negative stimuli? Pleasant stimuli usually tend to bring the stimulating source closer, and unpleasant stimuli tend to push it farther. In this experiment, the elderly worked in extremes of affective valence and alert level. However, those causing alert and a possible behavioral physical activity are unpleasant stimuli. Memories, learning and behavior tend to be better processed when faced with a stimulus and/or in situations that generate an alert status.^27,28 Some authors²⁹ state that it is not the pleasant or unpleasant aspect of a stimulus that favors memory, but the alerting aspect of this stimulus. Therefore, it is feasible that information learnt and required in the everyday to maintain body integrity in the elderly depends on the perfect recognition of what is considered to be unpleasant. The higher the alert, the higher the attention and the more reliable the created and learnt memory trait. In the elderly, the highest alert degree is reached by negative stimuli, such as seen in this study. Therefore, it is possible that, in geriatrics, memories and behaviors are guided by an accurate evaluation of displeasure, since it is this aspect that causes an alerting physiological response. A strong emotional focus seems to be basic for a geriatric cognitive evaluation.

Anatomically, integrities of amygdaloidal complexes, as well as other adjacent limbic areas, may also be involved in the "all-or-nothing" response by the elderly and are required in emotional evaluation.³⁰ However, the literature is controversial on this issue. Aging of the amygdalohippocampal and entorhinal complex seems to involute at a disproportional rhythm when compared with other brain regions, or not showing any signs of involution.²⁴ That is, there is evidence of aging in selectively sparing part of the limbic system and, on the other hand, compromising to a great extent the dorsolateral and orbitofrontal frontal systems.

CONCLUSIONS

In summary, this pilot project supports what has previously been suggested in the literature: how elderly individuals perceive and report emotion-loaded images can be different from that presented by young individuals, especially regarding alerting emotional stimuli. This observation points to the possibility of a tendency in considering an emotion more intensely than youths, relegating neutrality to the few stimuli of the environment and thus corroborating with the "all-or-nothing" aspect observed in the affective space diagram of this particular sample. What is neutral for the youths may not be for the elderly. Therefore, a definition of what pleasant, unpleasant and neutral are, and especially of what causes alert and relaxation in an elderly population should be well explained before evaluating emotion and attention and accessing emotional memory in this group. A normalization of the IAPS/SAM instruments is an essential condition for that.

It should also be considered that the results reported above are strongly restricted by statistical limitations. It is clear that this is a small sample, disproportional as to gender and restricted to a specific national region. However, it is known that a national normalization for the elderly is urgently necessary, and this is already an ongoing process.

REFERENCES

• 1. Pickholtz JL, Malamut BL. Cognitive changes associated with normal aging. In: Sirven JI, Malamut BL. Clinical neurology of the older adult. Philadelphia: Lippincott Williams & Wilkins; 2002. p. 56-64.
• 2. Denburg NL, Buchanan TW, Tranel D, Adolphs R. Evidence for preserve emotional memory in normal elderly persons. Emotion. 2003;3(3):239-54.
• 3. Mather M, Johnson MK. Choice-supportive source monitoring: Do our decisions seem better to us as we age? Psychol Aging. 2000;15(4):596-606.
• 4. Mather M, Carstensen LL. Aging and attentional biases for emotional faces. Psychol Sci. 2003;14(5):409-15.
• 5. Charles ST, Mather M, Carstensen LL. Aging and emotional memory: the forgettable nature of negative images for older adults. J Exp Psychol Gen. 2003;132(2):310-24.
• 6. Keefover RW. Aging and cognition. Neurol Clin. 1998;16(3):635-48.
• 7. Mather M, Johnson MK. Affective review and schema reliance in memory in older and younger adults. Am J Psychol. 2003;116(2):169-89.
• 8. Mather M, Canli T, English T, Whitfield S, Wais P, Ochsner K, et al. Amygdala responses to emotionally valenced stimuli in older and younger adults. Psychol Sci. 2004;15(4):259-63.
• 9. Bradley MM, Lang PJ. Measuring emotion: Behavior, feeling and physiology. In: Lane R, Nadel L, eds. Congitive neuroscience of emotion. New York: Oxford University; 2000. p. 242-73.
• 10. Bradley MM, Greenwald MK, Petry MC, Lang PJ. Remembering pictures: pleasure and arousal in memory. J Exp Psychol Learn Mem Cogn. 1992;18(2):379-90.
• 11. Bradley MM, Lang PJ. Measuring emotion: the self-assessment manikin and the semantic differential. J Behav Ther Exp Psychiatry. 1994;25(1):49-59.
• 12. Cook EW 3rd, Melamed BG, Cuthbert BN, McNeil DW, Lang PJ. Emotional imagery and the differential diagnosis of anxiety. J Consult Clin Psychol. 1988;56(5):734-40.
• 13. Baddeley AD. Human memory: theory and practice. Allyn & Bacon; 1998. p. 273-92.
• 14. Cahill L. Neurobiology of memory for emotional events: converging evidence from ingra-human and human studies. Cold Spring Harb Symp Quant Biol. 1996;61:259-64.
• 15. Lang PJ, Bradley MM, Cuthbert BN. International Affective Picture System (IAPS): Instruction Manual and Affective Ratings. Gainesville: University of Florida; 1999.
• 16. Moltó J, Montañés S, Poy R, Segarra P, Pastor MC, Tormo MP, et al. Un nuevo método para el estudio experimental de las emociones: el internacional affective picture system (IAPS). Adaptación española. Rev Psicol Gen Aplicada. 1999;52(1):55-87.
• 17. Ribeiro RL. Avaliação de relatos subjetivos e de alterações fisiológicas a estímulos do "International Affective Picture System" (IAPS) em estudantes universitários brasileiros (tese). São Paulo: Universidade Federal de São Paulo, Escola Paulista de Medicina; 2003.
• 18. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9(3):179-86.
• 19. Hachinski VC, Iliff LD, Zilhka E, Du Boulay GH, McAllister VL, Marshall J, et al. Cerebral blood flow in dementia. Arch Neurol. 1975;32(9):632-7.
• 20. Freedman M, Leach L, Kaplan EF, Winocour G, Shulman KI, Deli DC. Clock drawing: a neuropsychological analysis. New York: Oxford University; 1994.
• 21. Okamoto IH. Aspectos cognitivos da doença de Alzheimer no teste do relógio: Avaliação de amostra da população brasileira (tese). São Paulo: Universidade Federal de São Paulo, Escola Paulista de Medicina; 2001.
• 22. Porto WG. Normatization of 702 images set from the International Affective Picture System (IAPS) in a Brazilian elderly sample for emotion and attention testing. Arq Neuro-Psiquiatr. 2006;64(1):162-3.
• 23. Cuthbert BN, Bradley MM, Lang PJ. Psychophysiological responses to affective slides across the life span. Psychophysiology. 1988;25(4):441.
• 24. Mather M. Aging and emotional memory. Reisberg D, Hertel P, eds. Memory and emotion. New York: Oxford University. 2003. p. 272-307.
• 25. Carstensen LL, Charles ST. Emotion in the second half of life. Cur Directions Psychol Sci. 1998;7(5):144-49.
• 26. Charles ST, Mather M, Carstensen LL. Aging and emotional memory: the forgettable nature of negative images for older adults. J Exp Psychol Gen. 2003;132(2):310-24.
• 27. Izquierdo I. Memória. Porto Alegre: Artmed. 2002. p. 9-95.
• 28. LeDoux J. O cérebro emotional. Santos TB. Rio de Janeiro: Objetiva. 1998. p. 9-332.
• 29. Bradley MM, Greenwald MK, Petry MC, Lang PJ. Remembering pictures: pleasure and arousal in memory. J Exp Psychol. 1992;18(2):379-90.
• 30. Bradley MM, Lang PJ. Measuring emotion: the self-assessment manikin and the semantic differential. J Behav Ther Exp Psychiatry. 1994;25(1):49-59.

Publication Dates

• Publication in this collection
  06 Jan 2009
• Date of issue
  Aug 2008

History

• Accepted
  11 Apr 2008
• Received
  27 Nov 2007