Farry, R.
2026 | Ergonomics and Human Factors 2026 (forthcoming)
Adversarial Design Thinking for Organisational Architecture
Organisations can unintentionally create friction, dysfunction, and harm through the design of their structures, processes, and information flows. This paper introduces Adversarial Design Thinking, a parallel‐design method that applies a malicious‐insider mindset to organisational architecture to reveal these hidden vulnerabilities. A Red Team is tasked to design solutions that meet stated goals while maximising plausible, undetected organisational harm, while a Blue Team designs conventionally. Comparing their outputs surfaces latent risks, structural weaknesses, and unintended consequences that human‐centred approaches—often assuming good intent—may overlook. The paper presents the GHOST and Harm frameworks to support identification of adversarial design patterns, showing how organisational features can hide harm, degrade recovery, and allow dysfunction to accumulate. This lens strengthens organisational resilience and design quality.
More info on ADT >
Farry, R.
2025 | PhD thesis paper – The Misled Mind
Where is my Mind?
In this paper I argue in favour of radical cognitive extension, sometimes referred to as cognitive bloat. According to the Extended Mind Thesis (Clark and Chalmers 1998), under the right conditions cognition and the mind can extend over external artefacts like notebooks. Various critics—along with Clark and Chalmers 1998—seek to limit the scope of the thesis, often appealing to a reductio ad absurdum. However, rather than the implications of the Extended Mind Thesis being absurd, I claim they are counter-intuitive but revealing insights into the nature of cognition, and processes in general. In this paper I argue against Clark and Chalmers’ (1998) ‘Availability and Portability’ Criteria and consider a parallel case of Extended Digestion and the extent to which we consider entities apt for a particular functional role is based on human-centric concerns, to make the case for a wide view of cognitive extension, in favour of cognitive bloat.
Farry, R.
2025 | PhD thesis paper – The Misled Mind
In Defence of the Interpretative Sensory-Access Theory of Self-Knowledge: confabulation does undermine introspection for propositional attitudes
Knowing thyself, through a glass, darkly
Cases of the confabulation of propositional attitudes have been used by Carruthers (2013) as evidence that we lack introspective access to our propositional attitudes, and that instead our route to self-knowledge is via self-interpretation. In this paper I defend this position from objections raised by Andreotta (2021) that Carruthers’ claimed cases of self-misattributions do not show people misattributing their propositional attitudes. In response to Andreotta’s objection for a particular case involving a split-brain patient, that the apparent misattribution involved a fabricated explanation, not a fabricated intention, I argue that this objection does not hold for cases where self-attributions are made retrospectively. Secondly, I explore the unwelcome consequences of accepting a separate objection from Andreotta that some claimed cases of misattributions are instances of propositional attitudes being influenced by perceptual cues. Accepting this objection would entail that we are mercurial beings, which plainly we are not, or that additional attitude-like mental features would be required to avoid this entailment. In defending self-interpretation accounts of self-knowledge, I seek to show that compared to direct access views they can more plausibly account for why our self-reports are swayed by perceptual cues. I go on to make the case that confabulations are a normal and everyday occurrence, and as such, our confabulations mainly go unnoticed, get glossed over, or become self-fulfilling. Consequently, the level of behavioural incongruence with self-attributions required by critics to show that confabulations have occurred is too demanding.
Farry, R.
2025 | PhD thesis paper – The Misled Mind
Useful But Not Accurate: an argument for illusionism
We experience things like redness and painfulness—the qualitative aspects of experience—which seem difficult to account for in terms of the brain and information processing. This difficulty arises, because the qualitative aspects of experience seem to have features which (when taken at face value) seemingly do not fit comfortably with a physicalist and mechanistic view of the brain. There are several different approaches to resolving this difficulty, including Keith Frankish’s (2017; 2023) Illusionism. According to Illusionism we only seem to experience things like redness and painfulness. They do not really exist, rather they are the result of a misrepresentation or illusory perspective, somewhat analogous to the apparent existence of rainbows as arcs in the sky.
In this paper I present a novel argument in support of Illusionism, that as evolved entities we should expect our various mental capacities to track usefulness rather than accuracy. As such the way things seem to us seem the way that they do because they have developed to be useful, rather than truth-tracking, and so we should not expect our qualitive experiences to provide us with a view of how things really are. The various perceptual illusions and cognitive biases that we experience show that in a range of cases such expectations are correct. As a result, we should be wary of using our qualitative experiences as a basis for determining what things really are, and in particular the metaphysical status of those experiences.
Hillyer, C., State-Davey, H., Hooker, N., Farry, R., Bond, R., Campbell, J., Morgan, P., Jones, D., Vega, J., and Butler, P.
2023 | Ergonomics and Human Factors 2023
Human Factors Guidance for Robotic and Autonomous Systems (RAS)
This paper outlines recent (2021/2022) work to produce Human Factors (HF) guidance to support the design, development, evaluation, and acquisition of Robotic and Autonomous Systems.
Farry, R.
2023 | Ergonomics and Human Factors 2023
Using SUS for Current and Future AI
The System Usability Scale (SUS) was assessed for its relevance and ease of use for assessing an AI capable of human-like interaction. Participants used SUS to assess Outlook, a contemporary consumer-grade AI interaction partners (smartphone digital assistants), and human teammates as a proxy ‘system’ for future human-like AI interaction partners. The results show that participants considered SUS to be relevant and easy to use for contemporary consumer-grade AI interaction partners, but not for human teammates. However, there was no meaningful difference in their ability to apply SUS between contemporary digital assistants, human teammates, and an email client. Thus, SUS can be used effectively for all of these kinds of systems.
Farry, R.
2020 | Ergonomics and Human Factors 2020
Predicting how people will respond to a disruptive event: The human factors response framework
System disruptions can have far reaching negative consequences. The extent to which a system can anticipate, absorb and adapt to a disruption is a characteristic of its resilience. As people are often fundamental to system resilience, an improved understanding of the people-related factors that underpin system resilience helps in predicting system vulnerability and the response to a disruptive event. The Human Factors Response Framework was developed to provide this improved understanding. The framework supports analysts in identifying relevant people-related factors within a system, and the prediction of the system’s resilience and the likely dominant response from key personnel. This paper provides a high-level overview of the framework, its development, and future research direction.
