A Comparison of Younger and Older Adults' Self-Assessment Manikin Ratings of Affective Pictures

ABSTRACT

Self-Assessment Manikin (SAM) ratings were used to examine whether groups of 21 younger (M age = 20.02 years, SD = 2.28) and 21 older (M age = 66.26 years, SD = 5.64) adults had similar affective experiences to pictures from the International Affective Picture System (Lang, Bradley, & Cuthbert, 1997). The psychometrics of the SAM valence and arousal scales were also compared across age groups. Internal consistency reliability (Cronbach's alpha) was similar for younger and older adults, where both groups made less consistent valence ratings than arousal ratings. Both groups differed from the norms for valence for pleasant pictures, but were no more different from each other than they were from the norms. Age group differences were most evident in the pleasant region of the bivariate valence by arousal affective space, where younger adults found pleasant-aroused pictures to be more pleasant and arousing than older adults did. We suggest that this age group difference could be explained by greater affect intensity and surgency for the younger group and greater emotional control and leveling of positive affect for the older group.

This research was supported by a Central Michigan University Faculty Research and Creative Endeavors grant to the first author. The authors thank Jamie Sicard for her assistance with programming and data collection.

Notes

¹The pleasant-aroused region of the IAPS includes pictures of male and female nudes and erotic pictures of couples. Participants are often shown opposite sex nudes, which require that different picture sets from this region be used for males and females. We used only erotic pictures that included couples so that the same pleasant-aroused pictures were shown to all participants.

Note. PA: pleasant-aroused; PC: pleasant-calm; N: neutral; UA: unpleasant-aroused; UC: unpleasant-calm. SAM valence and arousal scales range from 1 to 9. The unit of analysis was the 18 pictures in each affect region. d ₁: Standardized mean difference between norm and younger groups. d ₂: Standardized mean difference between norm and older groups. d ₃: Standardized mean difference between younger and older groups.

^a Lang et al. (1997).

^b t test for age group comparison p < .05.

Note. Each affect region has 18 pictures.

*p < .05;

**p < .01.

²For the five levels of affect region in arousal, the following four orthogonal coding vectors were created: (a) contrast between the average of pleasant- and unpleasant-aroused regions and the average of pleasant- and unpleasant-calm regions; (b) contrast between pleasant-aroused and unpleasant-aroused; (c) contrast between pleasant-calm and unpleasant-calm; (d) contrast between neutral and the average of the other four affect regions. For the five levels of affect region in valence, the following four orthogonal coding vectors were used: (a) contrast between the average of pleasant-aroused and pleasant-calm regions and the average of unpleasant-aroused and unpleasant-calm regions; (b) contrast between pleasant-aroused and pleasant-calm; (c) contrast between unpleasant-aroused and unpleasant-calm; (d) contrast between neutral and the average of the other four affect regions. The age group by affect region interaction term was created by multiplying the age vector by each of four orthogonally coded affect region vectors. There were four interaction terms each for the arousal and for valence dimensions.

Additional information

Notes on contributors

Richard W. Backs

The current address of Sergio da Silva is Department of Psychology, Calvin College, Grand Rapids, Michigan, USA.