Verified by 
As I pointed out in a previous post, there is now good evidence from brain scanning that mentalism (thinking about things in mental terms) recruits different circuits in the brain than mechanistic cognition (thinking about things in physical terms). According to the diametric model of mental illness, autism spectrum disorders are the result of severe deficits in mentalizing (hypo-mentalism), while psychotic spectrum disorders are the opposite: the result of pathological excesses in mentalizing (hyper-mentalism).
As an earlier post noted, evidence of this has already been found in the brain. But now a remarkable experiment carried out by Ahmad Abu-Akel and colleagues has been “the first to show that the postulated diametric modulation of autism tendencies and psychosis proneness on behavior and performance are detectable at the neural level in a region that is a core component of social functioning.”
The study examined the concurrent effect of autism traits and psychosis-proneness on the activity of the mentalizing network in 24 neuro-typical adults performing a competitive social game: the Rock, Paper, Scissors (RPS) task while undergoing a brain scan. In the RPS task, participants are required to predict the moves of their opponent in order to win. The game has the following simple rules: Rock beats scissors, paper beats rock, and scissors beats paper. The winner of each round is awarded 1 point. A no-response results in an automatic win for the opponent and identical moves results in a draw and no points are awarded.
The researchers led the participants to believe that they were playing under four conditions: (1) against an active human agent who was a skilled RPS player; (2) a passive human agent who followed a pre-determined script; (3) an active intelligent computer program that was capable of analyzing the participant’s strategy; and (4) a passive computer program that followed a pre-determined script.
Participants were cautioned not to use a rigid strategy and to play competitively. The highest performing participant at the end of the study got a prize of ₤10. But unbeknown to the participants, the game was always played against a computer program generating moves entirely at random. The design ensured that the only difference across the conditions was the perceived identity of the participant’s opponent. None of the participants expressed any doubt regarding the identity of the four opponents when debriefed after the scanning sessions.
Activations were observed in the paracingulate cortex and the right temporo-parietal junction (rTPJ) illustrated left. Autism traits and psychosis proneness did not modulate activity within the paracingulate or the dorsal component of the rTPJ.
However, diametric modulations of autism traits and psychosis proneness were observed in the posterior (rvpTPJ) and anterior (rvaTPJ) subdivisions of the ventral rTPJ, which respectively constitute core regions within the mentalizing and attention-reorienting networks illustrated above. Within the rvpTPJ, increasing autism tendencies decreased activity, and increasing psychosis proneness increased activity. This effect was reversed within the rvaTPJ.
According to the researchers and as illustrated by them above:
The results indicated that autism tendencies and psychosis proneness have diametric influences on the neural activity within the ventral posterior (mentalizing) and anterior (attention-reorienting) subdivisions of the rTPJ. Specifically, while autism tendencies were associated with decreased activity in the ventral posterior rTPJ, psychosis proneness was associated with increased activity. Intriguingly, this pattern was reversed for the ventral anterior subdivision of the rTPJ, such that activity was positively associated with autism tendencies and negatively with psychosis proneness. Contrary to our expectations, task-related activations within the paracingulate cortex were unrelated to interindividual differences in autism tendencies or psychosis proneness. While this null finding may simply be due to not having sufficient power, an intriguing possibility for future research is to examine whether autism and psychosis expressions affect activity of posterior regions within the mentalizing network, which are involved in the representation of mental states, differently than anterior regions, which are more involved in the application and deployment of represented mental states (…).
The researchers add that “The nature of the interactive effect of autism and psychosis expressions on rTPJ activity is consistent with the diametric model positing that autism and schizophrenia spectrum disorders are etiologically and phenotypically diametrical exerting opposing influences on activity and behavior,” as proposed by the diametric model. Furthermore, it is worth adding that these predicted diametrically-opposite effects were seen in the brains of normal subjects, not clinically-diagnosed cases. This strikingly confirms the model's implication that autism and psychosis are simply extreme expressions of completely normal mentalizing, which represents a balance of both.
Finally—and citing the imprinted brain theory—the authors conclude that:
Delineation of the causal links among subdivisions within the TPJ would be an important step forward in understanding their role within the mentalizing network. Furthermore, the opposite effects of autism and psychosis on neural activity within the TPJ suggest that these conditions influence independent yet interacting systems, which may be precipitated by discrete genetic mechanisms (Crespi & Badcock 2008).
Thanks to Ahmad Abu-Akel for his help with this post.
