Semantic feature dissociation: A new hypothesis concerning autism

ABSTRACT

In this paper, I outline a new candidate hypothesis concerning autism: Semantic Feature Dissociation (SFD). This is the claim that, in some cases of autism, connections between feature representations in semantic memory may be weaker. More specifically, connections representing low-strength correlations may be disproportionately lost. I demonstrate the wide-ranging effects this change would have by introducing two analytical categories, concept narrowing (CN) (a tendency to make fewer inferences from the same concepts) and concept specialization (CS) (a tendency to be sensitive to fewer cues in categorization). Presenting the results of a novel qualitative study of autism autobiographies, I show that SFD can plausibly explain many common autism traits, as described by autistic autobiographers. I conclude by considering how far this account might generalize, and how it might be related to existing theories of autism.

KEYWORDS:

• Autism
• autobiography
• semantic memory
• concepts
• categorization

Disclosure statement

No potential conflict of interest was reported by the author.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1. Asperger’s Syndrome (AS) no longer exists as a diagnosis in the DSM-5. Previously, it was defined by the absence of significant language impairment. It is safe to assume that all published autobiographers are linguistically competent (excluding some difficulties with pragmatics which would be consistent with AS). Hence, I do not distinguish between AS and autism here.

2. This example can also help highlight the relationship between the two descriptive tools, CN and CS. As I noted earlier, concepts are systematically overlapping and interrelated. So, while I interpret this example in terms of CN (as narrowing of a classroom schema), it can also be readily interpreted in terms of CS (not activating a question schema).

3. One alternative interpretation of Dumortier’s remark (helpfully pointed out by the reviewer), might be that she is speaking figuratively, i.e.: “I know this is salami, but it’s not proper salami: I refuse to eat this stuff.” However, Dumortier reports being unable, not just unwilling, to eat. Hence, the reading I prefer here is that she does not infer core properties of the salami, including that it is edible. As noted in the introduction, conceptual inferences are believed to be rapid, automatic, and closely tied up with affordances. Naturally, I assume Dumortier can understand what the salami is after reflection, or she would never have been able to interpret and write about the experience.

Additional information

Funding

This work was supported by an Arts and Humanities Research Council studentship awarded to the author through the CHASE Doctoral Training Partnership.

Notes on contributors

Ian Hare

Ian Hare is a philosophy PhD student at the University of East Anglia. His research focuses on philosophy of psychiatry and psychology, especially autism and ADHD.