arXiv subjects

Tom Owen

Publications and source records attributed to Tom Owen.

3 recordsLinked to original sources

arXiv2026

Every(bot) Makes Mistakes: Coding Big Five Personalities, Context, and Tone into an LLM Chatbot Recovery Code Framework

Despite careful design involving classifiers, parameters, and safeguarding, errors during human/AI interaction are not rare. Poor error recovery can disrupt interaction flow, damage user trust, and decrease user engagement. Whilst existing work has explored LLM recovery, tone, context, and personality as separate design dimensions, no existing work has combined these variables into a structured guidance framework. This paper presents a recovery code that maps four common LLM chatbot task contexts to associated personality traits (four Big Five personalities: Conscientiousness, Agreeableness, Openness, and Extraversion), tones, and three-stage recovery instructions. A recovery evaluation rubric was also designed, comprising three dimensions (Recovery quality, Tone alignment, and Appropriateness) and nine sub-dimensions. The methodology is exploratory, with no participants used. A between-subjects design was employed across two conditions: Condition A (baseline, uncoded), four separate Claude Sonnet 4.6 agents received no recovery code training; Condition B (coded), four separate Claude Sonnet 4.6 models were trained on the recovery code. Identical 'user' prompts and error scenarios were used across both conditions. Eight LLM evaluator agents assessed the recovery responses using the evaluation rubric, producing scores out of 5 for each sub-dimension. Results found a 27.8% average performance increase in coded recovery responses (76.7%) compared to baseline responses (48.9%). Condition B performed strongest in the appropriateness dimension (83.3%), with notable improvement in personality appropriateness (75% versus 50%) and providing explanation (60% versus 20%). These findings suggest that structured personality, context, and tone-informed recovery codes can be successfully learnt and applied by LLM chatbots to improve error recovery quality across varying contextual tasks.

cs.HC↗

arXiv2026

Translating Ethical Frameworks Into User-Centred Anti-Social Behaviour Interventions

In 2025 one million Anti-Social Behaviour (ASB) cases were recorded in England & Wales, impacting community cohesion. Statutory guidance presents punitive interventions that lack technological input and does not often root ethical frameworks within government system design. This work takes a novel approach in framing ASB intervention as a human-computer interaction problem by embedding an ethical framework into two digital designs, aiming to increase public responsibility and prevent ASB. The first design is extracted from UK public opinion research, the ethical themes include punitive proportionality, personalisation, and responsibility. The second are digital interventions that present a set of QR-based public reporting interfaces and a web-based ASB awareness course that precedes punitive escalation. Our methodology involves structured interviews and online surveys. Results positively evaluated the framework and QR interfaces. Such outcomes could inform the expansion of technological intervention utilisation that does not replace existing punitive approaches, but balances them.

cs.HC↗

arXiv2023

Identifying epileptogenic abnormality by decomposing intracranial EEG and MEG power spectra

Identifying abnormal electroencephalographic activity is crucial in diagnosis and treatment of epilepsy. Recent studies showed that decomposing brain activity into periodic (oscillatory) and aperiodic (trend across all frequencies) components may illuminate drivers of changes in spectral activity. Using iEEG data from 234 subjects, we constructed a normative map and compared this with a separate cohort of 63 patients with refractory focal epilepsy being considered for neurosurgery. The normative map was computed using three approaches: (i) relative complete band power, (ii) relative band power with the aperiodic component removed (iii) the aperiodic exponent. Corresponding abnormalities were also calculated for each approach in the separate patient cohort. We investigated the spatial profiles of the three approaches, assessed their localizing ability, and replicated our findings in a separate modality using MEG. The normative maps of relative complete band power and relative periodic band power had similar spatial profiles. In the aperiodic normative map, exponent values were highest in the temporal lobe. Abnormality estimated through the complete band power robustly distinguished between good and bad outcome patients. Neither periodic band power nor aperiodic exponent abnormalities distinguished seizure outcome groups. Combining periodic and aperiodic abnormalities improved performance, similar to the complete band power approach. Our findings suggest that sparing cerebral tissue that generates abnormalities in either periodic or aperiodic activity may lead to a poor surgical outcome. Both periodic and aperiodic abnormalities are necessary to distinguish patient outcomes, with neither sufficient in isolation. Future studies could investigate whether periodic or aperiodic abnormalities are affected by the cerebral location or pathology.

q-bio.NC↗