Artificial Intelligence and Medical Humanities

The use of artificial intelligence in healthcare has led to debates about the role of human clinicians in the increasingly technological contexts of medicine. Some researchers have argued that AI will augment the capacities of physicians and increase their availability to provide empathy and other uniquely human forms of care to their patients. The human vulnerabilities experienced in the healthcare context raise the stakes of new technologies such as AI, and the human dimensions of AI in healthcare have particular significance for research in the humanities. This article explains four key areas of concern relating to AI and the role that medical/health humanities research can play in addressing them: definition and regulation of “medical” versus “health” data and apps; social determinants of health; narrative medicine; and technological mediation of care. Issues include data privacy and trust, flawed datasets and algorithmic bias, racial discrimination, and the rhetoric of humanism and disability. Through a discussion of potential humanities contributions to these emerging intersections with AI, this article will suggest future scholarly directions for the field.

     

Early evidence affects later decisions: Why evidence accumulation is required to explain response time data

Models of decision making differ in how they treat early evidence as it recedes in time. Standard models, such as the drift diffusion model, assume that evidence is gradually accumulated until it reaches a boundary and a decision is initiated. One recent model, the urgency gating model, has proposed that decision making does not require the accumulation of evidence at all. Instead, accumulation could be replaced by a simple urgency factor that scales with time. To distinguish between these fundamentally different accounts of decision making, we performed an experiment in which we manipulated the presence, duration, and valence of early evidence. We simulated the associated response time and error rate predictions from the drift diffusion model and the urgency gating model, fitting the models to the empirical data. The drift diffusion model predicted that variations in the evidence presented early in the trial would affect decisions later in that same trial. The urgency gating model predicted that none of these variations would have any effect. The behavioral data showed clear effects of early evidence on the subsequent decisions, in a manner consistent with the drift diffusion model. Our results cannot be explained by the urgency gating model, and they provide support for an evidence accumulation account of perceptual decision making.     

Dissociations between vision for perception and vision for action depend on the relative availability of egocentric and allocentric information

In three experiments, we scrutinized the dissociation between perception and action, as reflected by the contributions of egocentric and allocentric information. In Experiment 1, participants stood at the base of a large-scale one-tailed version of a Müller-Lyer illusion (with a hoop) and either threw a beanbag to the endpoint of the shaft or verbally estimated the egocentric distance to that location. The results confirmed an effect of the illusion on verbal estimates, but not on throwing, providing evidence for a dissociation between perception and action. In Experiment 2, participants observed a two-tailed version of the Müller-Lyer illusion from a distance of 1.5 m and performed the same tasks as in Experiment 1, yet neither the typical illusion effects nor a dissociation became apparent. Experiment 3 was a replication of Experiment 1, with the difference that participants stood at a distance of 1.5 m from the base of the one-tailed illusion. The results indicated an illusion effect on both the verbal estimate task and the throwing task; hence, there was no dissociation between perception and action. The presence (Exp. 1) and absence (Exp. 3) of a dissociation between perception and action may indicate that dissociations are a function of the relative availability of egocentric and allocentric information. When distance estimates are purely egocentric, dissociations between perception and action occur. However, when egocentric distance estimates have a (complementary) exocentric component, the use of allocentric information is promoted, and dissociations between perception and action are reduced or absent.     

Effects of Interpretation Training on Hostile Attribution Bias and Reactivity to Interpersonal Insult

Research suggests that individuals high in anger have a bias for attributing hostile intentions to ambiguous situations. The current study tested whether this interpretation bias can be altered to influence anger reactivity to an interpersonal insult using a single-session cognitive bias modification program. One hundred thirty-five undergraduate students were randomized to receive positive training, negative training, or a control condition. Anger reactivity to insult was then assessed. Positive training led to significantly greater increases in positive interpretation bias relative to the negative group, though these increases were only marginally greater than the control group. Negative training led to increased negative interpretation bias relative to other groups. During the insult, participants in the positive condition reported less anger than those in the control condition. Observers rated participants in the positive condition as less irritated than those in the negative condition and more amused than the other two conditions. Though mediation of effects via bias modification was not demonstrated, among the positive condition posttraining interpretation bias was correlated with self-reported anger, suggesting that positive training reduced anger reactivity by influencing interpretation biases. Findings suggest that positive interpretation training may be a promising treatment for reducing anger. However, the current study was conducted with a non-treatment-seeking student sample; further research with a treatment-seeking sample with problematic anger is necessary.     

The Hepatitis C treatment experience: Patients’ perceptions of the facilitators of and barriers to uptake,adherence and completion

Objective: This study explores the perceptions of patients receiving treatment for Hepatitis C to determine what factors influence their decision to commence treatment, ability to maintain adherence and complete their treatment program.

Design: Semi-structured interview techniques were used in a qualitative study of 20 patients undergoing treatment for Chronic Hepatitis C (CHC).

Main outcome measures: To explore patients’ perceived barriers and facilitators of Hepatitis C treatment adherence and completion.

Results: Analysis of patient interviews identified four key themes: (1) motivations for commencing CHC treatment – fear of death and ridding themselves of stigma and shame; (2) the influential role of provider communication – patients reported that information and feedback that was personalised to their needs and lifestyles was the most effective for improving adherence to treatment; (3) facilitators of treatment adherence and completion – social, emotional and practical support improved adherence and completion, as did temporarily ceasing employment; (4) barriers to treatment adherence and completion – these included side effects, stigma, a complicated dosing schedule and limitations of the public healthcare system.

Conclusion: To increase treatment adherence and completion rates, a patient-centred approach is required that addresses patients’ social, practical, and emotional support needs and adaptive coping strategies.     

Distress and emotional well-being in breast cancer patients prior to radiotherapy: an expectancy-based model

Understanding precursors to distress and emotional well-being (EWB) experienced in anticipation of radiotherapy would facilitate the ability to intervene with this emotional upset (i.e. higher distress, lower EWB). Thus, this study tested an expectancy-based model for explaining emotional upset in breast cancer patients prior to radiotherapy. Women affected by breast cancer (N?=?106) were recruited and participants completed questionnaires prior to commencing radiotherapy. Structural equation modelling was used to test a cross-sectional model, which assessed the ability of dispositional optimism (Life Orientation Test-Revised – two factors), response expectancies (Visual Analog Scale items), medical (type of surgery, cancer stage and chemotherapy history) and demographic (age, race, ethnicity, education and marital status) variables to predict both EWB (Functional Assessment of Chronic Illness Therapy – Emotional Well-being Subscale) and distress (Profile of Mood States – short version). The model represented a good fit to the data accounting for 65% of the variance in EWB and 69% in distress. Significant predictors of emotional upset were pessimism, response expectancies, Latina ethnicity, cancer stage and having had a mastectomy. These variables explained a large portion of emotional upset experienced prior to radiotherapy for breast cancer and are important to consider when aiming to reduce distress and improve EWB in this context.     

Using Clinician-Rated Five-Factor Model Data to Score the DSM–IV Personality Disorders

Proposals suggest that many or all of the Diagnostic and Statistical Manual of Mental Disorders (4th ed. [DSM–IV]; American Psychiatric Association, 1994) personality disorders (PDs) may be omitted from the DSM (5th ed.; DSM–V) and replaced with a dimensional trait model of personality pathology (Krueger, Skodol, Livesley, Shrout, &; Huang, 2007 Krueger, R. F., Skodol, A. E., Livesley, W. J., Shrout, P. E. and Huang, Y. 2007. Synthesizing dimensional and categorical approaches to personality disorders: Refining the research agenda for DSM-V Axis II. International Journal of Methods in Psychiatric Research, 16: 65–73. [Crossref], [PubMed], [Web of Science ®] , [Google Scholar]; Skodol, 2009 Skodol, A. 2009. Emerging DSM-V perspectives on personality disorders. Plenary address presented at the Eleventh International ISSPD Congress. August2009, New York, New York.  [Google Scholar]). Several authors have expressed concerns that this may be difficult for clinicians and researchers who are more comfortable with the extant PD diagnoses. In this study, we tested whether clinician ratings of traits from the Five-factor model (FFM; Costa &; McCrae, 1990) can be used to recreate DSM–IV PDs. Using a sample of 130 clinical outpatients, we tested the convergent and discriminant validity of the FFM PD counts in relation to consensus ratings of the DSM–IV PDs. We then examined whether the FFM and DSM–IV PD scores correlate in similar ways with self-reported personality traits from the Schedule for Nonadaptive and Adaptive Personality (Clark, 1993 Clark, L. A. 1993. Manual for the Schedule for Nonadaptive and Adaptive Personality (SNAP), Minneapolis: University of Minnesota Press.  [Google Scholar]). Finally, we tested the clinical utility of the FFM PD counts in relation to functional impairment. Overall, the FFM PD counts, scored using clinician ratings of the FFM traits, appeared to function like the DSM–IV PDs, thus suggesting that the use of a dimensional trait model of personality in the DSM–V may still allow for an assessment of the DSM–IV PD constructs.     

Time to act and attend to the real mechanisms of action and attention

We discuss Humphreys' article in the context of two challenges that exist in regards to future research on the link between action and attention: (1) determining the cognitive and neural mechanisms responsible for an action‐attention link and (2) demonstrating that the action‐attention links observed in the laboratory reflect the same links between action and attention observed in the complexities of everyday life.     

Decision making under risk condition in patients with Parkinson's disease: a behavioural and fMRI study

We aimed to study whether previously described impairment in decision making under risky conditions in patients with Parkinson's disease (PD) is affected by deficits in using information about potential incentives or by processing feedback (in terms of fictitious gains and losses following each decision). Additionally, we studied whether the neural correlates of using explicit information in decision making under risk differ between PD patients and healthy subjects. We investigated ten cognitively intact PD patients and twelve healthy subjects with the Game of Dice Task (GDT) to assess risky decision making, and with an fMRI paradigm to analyse the neural correlates of information integration in the deliberative decision phase. Behaviourally, PD patients showed selective impairment in the GDT but not on the fMRI task that did not include a feedback component. Healthy subjects exhibited lateral prefrontal, anterior cingulate and parietal activations when integrating decision-relevant information. Despite similar behavioural patterns on the fMRI task, patients exhibited reduced parietal activation. Behavioural results suggest that PD patients' deficits in risky decision making are dominated by impaired feedback utilization not compensable by intact cognitive functions. Our fMRI results suggest similarities but also differences in neural correlates when using explicit information for the decision process, potentially indicating different strategy application even if the interfering feedback component is excluded.     

Reflecting on the action or its outcome: Behavior representation level modulates high level outcome priming effects on self-agency experiences

Recent research suggests that one can have the feeling of being the cause of an action’s outcome, even in the absence of a prior intention to act. That is, experienced self-agency over behavior increases when outcome representations are primed outside of awareness, prior to executing the action and observing the resulting outcome. Based on the notion that behavior can be represented at different levels, we propose that priming outcome representations is more likely to augment self-agency experiences when the primed representation corresponds with a person’s behavior representation level. Three experiments, using different priming and self-agency tasks, both measuring and manipulating the level of behavior representation, confirmed this idea. Priming high level outcome representations enhanced experienced self-agency over behavior more strongly when behavior was represented at a higher level, rather than a lower level. Thus, priming effects on self-agency experiences critically depend on behavior representation level.