Information gathering in autism and alexithymia
This is a long post regarding an article I found intriguing. Lee, Long, Catmur, Hauser & Bird (2024) published “Information Gathering: Dissociable effects of autistic and alexithymic traits in youths aged 6–25 years” in Emotion. Here are highly edited excerpts:
Autistic youths tend to react negatively to uncertain events. Little is known about the cognitive processes associated with this intolerance of uncertainty, most notably the tendency to actively gather information to minimize uncertainty. Past research has relied on self-report measures that may not allow investigation of the multifaceted processes associated with intolerance of uncertainty, including information gathering. Alexithymia (difficulties in identifying and describing one’s own emotions) commonly co-occurs with autistic traits, but its role in information gathering has rarely been considered. Accordingly, 97 typically developing youths (aged 6–25 years) performed an information gathering task in which they were asked to gather information to infer socioemotional (emotional state) and nonsocial (clothing preference) information about another person when information gathering was costly versus not costly. Dimensional autistic traits were associated with more information gathering regardless of costs and information type. Computational modeling suggested this may be because of the delayed emergence of subjective costs of information gathering in high autistic trait individuals, resulting in later guesses.
The substantial heterogeneity present in the cognitive profile of individuals in the autistic population is well recognized. Recent clinical evidence suggests that cognitive atypicalities experienced by some autistic individuals not only impact core autistic symptoms but also exacerbate co-occurring mood symptoms and emotional distress, including anxious and depressive symptoms and irritability.
One cognitive difference associated with autistic traits is an increased need for information, a need that is thought to reflect intolerance of uncertainty (IoU)—a disproportionately negative or maladaptive reaction to situations and events characterized by uncertainty. IoU predicts elevated anxiety in autistic youths and mediates the relationship between autistic symptom severity and anxiety and internalizing symptoms. Recent debates point to the multifaceted nature of IoU, which includes (but is not limited to) the tendency to seek certainty and/or the feeling of being “paralyzed” in the presence of uncertainty. While IoU is not defined as a need for information gathering, individuals who are averse to uncertain situations may actively engage in information gathering as a coping mechanism to minimize uncertainty. Excessive information gathering (i.e., when information gathering ceases to have informational value or becomes overly costly) may therefore constitute a maladaptive response to IoU. Studying information gathering in response to uncertainty and how it is affected by autistic traits may therefore prove to be of clinical relevance.
Certain aspects of information gathering are assessed using the Intolerance of Uncertainty Scale (Carleton et al., 2007). However, information gathering behaviors are not measured independently from other aspects of IoU on this scale, and it is unclear if information gathering items would constitute a valid and reliable set that can be analyzed on its own. The parent’s subjective evaluation of observable behaviors may also result in a reporting bias.
Another key research gap is the potential role of alexithymia in information gathering. Alexithymia is defined as the inability to identify and express one’s own emotions and is a subclinical condition that frequently co-occurs with autism. Notably, it has been consistently shown that alexithymia is more strongly associated with mood symptoms and socioemotional processing difficulties experienced by some autistic individuals (e.g., facial emotion perception, multidomain interoception, and empathy) than autism or autistic traits/symptoms alone. Measuring both autistic and alexithymic traits is therefore useful to inform their relative contributions to information gathering given its relevance to IoU and anxiety.
To address the above research gaps, the focus of this experimental study is to gain a deeper understanding of the relationship between autistic traits and cognitive processes associated with information gathering, and to identify whether alexithymia contributes to information gathering—an aim better served through the use of experimental tasks than existing self-report questionnaires. One such task (e.g., Clark et al., 2006) asks participants to determine the predominant internal color (yellow/blue) of a number of closed boxes by opening any number of those boxes under two conditions—(a) when boxes can be opened at no cost and (b) in a costly condition where points are deducted for each box opened. Note that while self-reported IoU predicts information gathering on this task (Hauser, Moutoussis, Dayan, et al., 2017), the number of boxes opened on this task is not a direct measure of IoU (there are several processes thought to influence the degree of information gathering before participants make a choice) but it allows the measurement of several parameters that are useful in characterizing the cognitive correlates of aversion to uncertainty and information gathering. One relevant response pattern is when participants continue to open boxes when the probability of one color being predominant approaches (or equals) 1. For example, if a participant has opened 13 yellow boxes in a grid of 25 boxes, yellow must be the predominant color as there are only a maximum of 12 boxes that remain closed. Even if all closed boxes are blue, yellow will be the predominant color. If participants continue to open boxes after finding 13 boxes of one color, then this may indicate an aversion to uncertainty surrounding the color of the closed boxes. In these tasks, the measure of information gathering is the number of boxes opened before a choice is made. Using this paradigm, autistic adults have been shown to gather more information (but not significantly so; Vella et al., 2018), and autistic adolescents have also been shown to gather more information in a similar paradigm (Brosnan et al., 2014).
Therefore, to investigate processes related to information gathering, we administered a modified version of the information gathering task in a youth sample aged 6–25 years while also measuring autistic and alexithymic traits. The information gathering task was modified such that, in addition to the manipulation of information gathering costs (with/without external costs) as in previous studies, the task was reframed such that participants were asked to make a guess about two types of information: another person’s clothing preference or their emotional state. The comparison between participants’ behavior when gathering nonemotional versus emotional information allowed it to assess whether there were any selective effects of alexithymia for emotional information. Following each round of information gathering, we asked participants to rate their own emotional experiences in response to the points they had lost or gained, which allowed for any effect of alexithymia on the consistency of emotional responses to be identified. The inclusion of children, adolescents, and young adults allows the examination of any developmental trends in the relationships between autistic and alexithymic traits and information gathering.
This experimental study investigated the effect of autistic and alexithymic traits on information gathering in typically developing youths (aged 6–25 years) using a modified information gathering task.
Supporting our hypothesis, we reported that youths with higher autistic traits had an overall increased propensity to gather information regardless of cost and information type when either the total number of draws or the number of draws in relation to the point of certainty was analyzed. Increased information gathering was evident regardless of whether the uncertain scenario required the gathering of emotional or nonemotional information, suggesting a general tendency. Computational modeling suggested that increased information gathering may stem from the fact that the internal costs of information gathering arise later for those with higher autistic traits, which promotes a delayed decision regardless of costs and information type. Given the overall total number of same-type draws made, even in those with high autistic traits, it is clear that any IoU experienced by these individuals was not severe enough to make them consistently persist in making draws beyond the point of certainty. Having a greater tendency to gather information per se does not constitute an aversion to uncertainty; instead, it could reflect, for example, increased curiosity or a desire to make a better-informed decision, although it did result in a lower reward rate (when controlling for parent-reported alexithymia). Of course, a tendency for increased information gathering may be pathological if it is due to, or results in, an IoU. An increased tendency to gather information on related tasks has been found in those with anxiety and obsessive–compulsive symptoms, both of which are prevalent in the autistic population.
Alexithymia was related to a reduced tendency to gather information about another person’s emotional state when analyzing parent reports of youth alexithymia, although this association was not significant using self-report data. It is unclear how this relates to previous reports of a positive association between alexithymia and self-report IoU measures that also include information gathering tendencies in autistic individuals. Further work separating the many cognitive and affective components of IoU is necessary in order to gain a clear picture of the factors responsible for any increased tendency to experience IoU in alexithymia and autistic traits.
The discrepancy between results obtained using parent- and child-report measures of alexithymia is not unique to this study and calls for a methodological framework guiding the measurement of alexithymia in developmental populations. Some suggest that parent reports may be more reliable than child reports of alexithymic traits, but this hypothesis is yet to be tested empirically. Other potentially contributing factors are the parent’s ability to make reliable inferences about their child’s alexithymic traits (given that alexithymic children are more likely to have alexithymic parents; e.g., Szatmari et al., 2008) and contextual effects on parent–child interactions (including the presence of co-occurring psychiatric conditions in either the child, parent, or both).
Despite the above, both self- and parent-report measures of alexithymia revealed an association between alexithymia and the consistency of emotional responses to both real and imagined wins and losses. These findings add further support for the idea that alexithymia produces variability in the evaluation of emotional experiences, which is present in the context of information gathering. This would result in an increased degree of uncertainty in any situation with the potential to induce an emotional reaction in those with alexithymia, and therefore exacerbate any IoU.
In sum, the present study reported experimental evidence for the association between information gathering and autistic and alexithymic traits in a typically developing sample. In addition, alexithymia was shown to be associated with reduced consistency of emotional responses to both real and imagined wins and losses. Both findings are of relevance to IoU and demonstrate that fractionating IoU into its cognitive and affective components is likely to be of clinical relevance. Finally, further evidence of dissociable effects of autistic traits and alexithymia highlights the need to consider co-occurring alexithymia when examining emotion processing related to autistic traits.
I found this work intriguing, though they note that this work needs to be replicated with people diagnosed with autism. However, I loved the methodology (the article is available as full text).