ASD Is Characterised by Atypicalities in Emotion-Related Learning Processes: What Are the Implications?

Background: One of the most robust phenomena in the memory literature is that emotionally salient stimuli are remembered better and for longer than hedonically neutral stimuli. Interactions between the amygdala and the hippocampus are critical for mediating this effect and emotional influences on memory can be dissociated experimentally from other memory phenomena. For instance, Kensinger & Corkin (2004) showed that memories for emotionally arousing stimuli are selectively resistant to the effects of attentional manipulations at encoding. We have previously shown that the long-term retention of emotional material is compromised in ASD (Gaigg & Bowler, 2008) but it remains unclear whether the short-term influences of emotion on memory might be preserved (see Gaigg & Bowler, 2008 and South et al., 2008).

Objectives: Here we draw on the work of Kensinger & Corkin (2004) to test the prediction that the short-term retention of emotional stimuli in ASD is quantitatively preserved but mediated by domain-general memory processes rather than specific mechanisms that are engaged by arousing stimuli (i.e., amygdala-hippocampal interactions). Specifically, we predict that the encoding of arousing words in ASD is not immune to manipulations of attention, as is the case for typically developing (TD) individuals.

Methods: 30 TD and 30 ASD adults, matched on chronological age, gender and verbal ability, participated in the experiment. They were asked to try to remember two lists of words comprising emotionally arousing items (taboos & profanities), hedonically valenced but non-arousing items (e.g., accident, victory,...) and hedonically neutral but categorically related words (items of clothing). Words were presented one at a time on a PC monitor and participants were required to study one list under full-attention and one under divided-attention conditions. During the former, participants studied words without distractions whilst during the latter they were required to monitor a continuous sequence of tones for pitch changes. The order of conditions was counterbalanced and after each list participants wrote down all words they could recall.

Results: Groups performed similarly on the tone monitoring task (t(58) = .007, ns) and the free recall results confirmed our predictions. Thus, for the TD group the manipulation of attention affected recall of categorically related and valenced words significantly more than recall of arousing items (F(1,29) = 5.29, p < .05), confirming that qualitatively different processes are engaged during the encoding of arousing material. In the ASD group, by contrast, there was no indication of such an interaction between attention (full vs. divided) and word type (arousing vs. valenced vs. category) (F(1,29) = 0.33, ns).

Conclusions: Together with previous studies, our observations provide clear evidence that emotional learning mechanisms are compromised in ASD, which has significant implications for developmental theory. Most important amongst these is that it forces us to re-consider the wide-spread assumption that only the social-emotional functions of the amygdala are compromised in ASD. We will argue against social-motivational theories that are based on this assumption and suggest instead that the developmental trajectory of ASD is more fruitfully conceptualised as resulting from a disruption of basic emotion-related learning mechanisms.