Perspective Taking and Memory for Self- and Town-Related Information in Male Adolescents and Young Adults

Abstract

Adolescence is a sensitive period for categorical self-concept development, which affects the ability to take others' perspectives, which might differ from one's own, and how self-related information is memorized. Little is known about whether these two processes are related in adolescence. The current study recruited 97 male participants aged 11-35 years. Using a self-referential memory task, we found that younger participants were less prone to recognize previously seen town-related adjectives, compared to self-related adjectives. However, this age-related reduction in recognition bias was unrelated to accurate memory performance. Using the Director task to assess perspective taking, we found an age-related decrease in egocentric biases in perspective taking from adolescence to early adulthood (i.e., perspective taking abilities improved with age). However, there was no evidence that these two processes were related. Overall, our findings suggest that male adolescents display parallel but independent age-related changes in self-referential biases in memory and perspective taking.

Keywords: adolescence; perspective taking; self-concept; self-referential processing; social cognitive development; theory of mind.

Figure 1. Tasks

Note. The self-referential memory task (A-B). (A) Learning phase: participants had to decide whether randomly presented adjectives (20 self-related and 20 town-related) were good descriptions of either themselves (Self condition: “Does this word describe yourself?”) or London (Town condition: “Does this word describe London?”). Responses were given on an 11-point rating scale. (B) Recall phase: participants had to judge whether they have already seen one of the randomly presented adjectives (40 target adjectives, 80 distractor adjectives) during the learning phase or not. Responses were given on a 5-point rating scale. The Director task (C-D). In this example, participants were verbally instructed by the director to “move the large jar right”. (C) During experimental trials, an error would be committed when ignoring the director’s perspective and incorrectly moving the distractor object, which is not visible to the director. In contrast, a correct response would be to move the target object, which is visible to both director and participant. (D) During control trials, distractor objects were replaced with irrelevant objects (e.g., the duck).

Figure 3. Effect of Age on Memory Sensitivity and Response Bias in the Self-Referential Memory Task

Note. (A) Effect of age on memory sensitivity. Mean d’ (y-axis) is plotted as a function of age (x-axis) for each condition. There was no significant change with age in the efficiency of processing self (vs. town) related adjectives. (B) Effect of age on response bias. Mean c (y-axis) is plotted as a function of age (x-axis) for each condition. With increasing age, participants demonstrated a reduction in “conservative” response bias for town-related adjectives, which indicates that participants became more likely to report town-related adjectives as present during the learning phase (slope_town = -0.02, SE = .005, p_Bonf > .001). All colored lines and shaded .95 confidence intervals (CIs of the fixed effects) show the linear trends as estimated by the linear mixed-effects models (A & B). *** p_Bonf < .001.

Figure 4. Effect of Age on Perspective Taking Accuracy in the Director Task

Note. Mean percentage correct (y-axis) is plotted as a function of age (x-axis) for each trial type. During experimental trials, participants had to consider the director’s perspective to select the correct answer, while this was not necessary to correctly answer in control trials. With increasing age, participants demonstrated a greater ability to take account of another person’s perspective (slope _experimental = 0.07, SE = .02, p_Bonf = .01). All colored lines and shaded .95 CIs show the linear trends as estimated by the trial-level generalized linear mixed-effects model. ** p_Bonf < .01.