Smart Environments for Diverse Cognitive Styles: the Case of Autism

Abstract

This paper, drawing from complex systems theory (CST), advances a framework that proposes the implementation of Ambient Smart Environments (ASEs) for personalised cognitive styles. The paper begins by arguing that Autism Spectrum Condition (ASC) can be described and understood as embodied misattunement with a neurotypical (sociocultural) environment. Personalising the environment is an opportunity for fitting a cognitive style, e.g. autistic interaction with the environment. High variability, noise, and rigidity can result in difficulties adapting to (neurotypical) environmental demands. This may result in rigid or repetitive behaviour, corresponding, in CST, to a “stuck state” or “local minimum”. Preventive intervention with ASE (by increasing free energy and moving out of the “stuck state” and into a new stable state) for autistic interaction with the environment can help develop patterns of flexible adjustment and attunement to the environment. ASE offers an opportunity for personalised therapeutic intervention for manipulating the environment to fit a cognitive style.

Extending Minds, Shifting Boundaries: ambient technological environments as extended allostatic control

“…In these accounts, the process of niche construction is usually characterised as simplifying or constraining the variety or complexity of a cognitive space (Constant et al, 2022). On the other hand, and relevant for our purposes, recent work has argued that the ability to increase complexity could have therapeutic potential, impacting well-being and psychopathology by disrupting overly powerful attractor states (which are hypothesised to underpin various forms of psychopathology (White & Hipólito, 2023). To take a simple example, consider how an ASE could subtly destabilise the kinds of habitual avoidance behaviours that can entrench episodes of depression, such as becoming socially withdrawn, and relying on the same repetitive routines of physical and mental inactivity.…”

“…Recent work has brought into focus ambient smart environments (ASEs), highlighting the conceptual differences between such systems and the technology that has featured in most of the TXM literature. It's argued that ASEs exhibit a radical, unique, functionality that sees them playing an 'upstream' role in attention and action selection (e.g., Verbeek, 2009;Aydin et al, 2019;White & Miller, 2023;White & Hipólito, 2023). In this article, we assess the applicability of TXM to ASEs, arguing that ASEs present a challenging case for classical formulations of TXM.…”

“…A key example might be the aetiology of different kinds of health complaints, which are often difficult (or impossible) to track, given the kaleidoscopic array of potential causal factors in the modern world. The insight of such systems may well be far deeper than that of even the most perceptive and self-reflective individuals (or, indeed, human experts) (White & Miller, 2023;White & Hipólito, 2024;Hipólito & White, 2023). Current smart technologies, such as apps running on our smartphones, already monitor certain behaviours, learning what we like and what interests us.…”

“…But when the whole point of the functionality of the system is for the user to not have to do anything, it's difficult, if not impossible, to describe that system in terms of affordances (White & Miller, 2023). Arguing that an affordance-based framework is ill-suited when applied to ASEs, recent work has claimed that the best way to understand the functionality of ASEs is through the concept of a "meta-affordance" (White & Miller, 2023;White & Hipólito, 2023;Hipólito & White, 2023). Understood as a meta-affordance, ASEs are seen as directly influencing which affordances the agent is likely to pay attention to, actively sculpting the agent's field of affordances in real time (White & Miller, 2023).…”