Alexithymia Is Associated with Impaired Interoceptive Accuracy but Not Interoceptive Signal Perception in ASD and Other Populations

Background:

Alexithymia, a condition characterized by difficulties identifying and describing one’s own emotions, has high co-occurrence with ASD and other disorders (Bird & Cook, 2013). Prior research is mixed, with studies demonstrating that alexithymia is associated with both heightened (Scarpazza et al., 2015) and reduced (Herbert et al., 2011) perception of interoceptive signals (i.e., internal bodily cues and sensations). Both heightened and reduced signal perception can be maladaptive; the former leading to poor modulation of interoceptive signals, and the latter leading to inadequate detection of interoceptive signals. Maintaining healthy bodily and emotional functioning requires “accuracy” in discriminating interoceptive signals and associating those signals with specific feelings or physiological states (e.g., hungry, nauseous, or angry). Delineating the mechanisms that underlie alexithymia is critical for understanding shared psychophysiological risk factors for various disorders and for considering treatments for alexithymia.

Objectives:

To conduct a meta-analysis to determine the extent to which various facets of interoceptive ability and alexithymia are related, and to determine the methodological factors and sample characteristics that account for heterogeneity among studies.

Methods:

Meta-analysis aggregated effects from all empirical studies reporting a statistical association between interoceptive ability and alexithymia across typically developing and disordered populations. Interoceptive ability was conceptualized into subjective and objective dimensions aligning with orientation to and ability to discriminate/specify bodily cues: 1) Objective Signal Perception—the ability to perceive and track bodily processes, such as one’s own heartbeat, 2) Subjective Signal Perception—self-reported awareness of bodily cues, and 3) Interoceptive Accuracy—self-reported ability to discriminate bodily cues or associate them with feelings or physiological states. Moderating effects of Diagnosis—typically developing, ASD, and eating disordered–were also examined.

Results:

A random effects model revealed a summary effect size across all 55 independent samples (total N = 6246) of r=-.161, p=.004, indicating that interoceptive ability is associated with lower alexithymia across studies. Analyzing facets of interoceptive ability separately, neither Objective Signal Perception nor Subjective Signal Perception (r=.033, df=28, p=.514), was significantly related to alexithymia (r=.080, df=13, p=.352). However, Interoceptive Accuracy was related to lower alexithymia (r=-.452, df=18, p<.001). Combining all subgroups, Diagnosis significantly moderated effect size strength Q_B(2)=18.61,p<.001. While interoceptive ability was not significantly related to alexithymia in typically developing samples (r=-.024, df=37, p=.616), it was related to lower alexithymia in ASD (r=-.501, df=5 p=.006) and eating disordered samples (r=-.493, df=6, p<.001).

Conclusions:

Reduced interoceptive ability is associated with alexithymia across studies, and this effect is driven by reduced Interoceptive Accuracy but not Objective or Subjective Interoceptive Signal Perception. This finding clarifies confusion in the literature and improves conceptual understanding of the alexithymia construct. We propose that ASD may be associated with both hypo and hyper-perception of interoceptive cues (negating statistically significant associations between Signal Perception and alexithymia), but that impaired Interoceptive Accuracy accounts for higher alexithymia in ASD and other disorders. The moderator analysis suggests that the mechanisms contributing to alexithymia may differ according to diagnosis, such that interoceptive deficits underlie alexithymia in ASD and eating disorders but not in the general population.