Capacity of Generative AI to Interpret Human Emotions From Visual and Textual Data: Pilot Evaluation Study.

BACKGROUND: Mentalization, which is integral to human cognitive processes, pertains to the interpretation of one's own and others' mental states, including emotions, beliefs, and intentions. With the advent of artificial intelligence (AI) and the prominence of large language models in mental health applications, questions persist about their aptitude in emotional comprehension. The prior iteration of the large language model from OpenAI, ChatGPT-3.5, demonstrated an advanced capacity to interpret emotions from textual data, surpassing human benchmarks. Given the introduction of ChatGPT-4, with its enhanced visual processing capabilities, and considering Google Bard's existing visual functionalities, a rigorous assessment of their proficiency in visual mentalizing is warranted. OBJECTIVE: The aim of the research was to critically evaluate the capabilities of ChatGPT-4 and Google Bard with regard to their competence in discerning visual mentalizing indicators as contrasted with their textual-based mentalizing abilities. METHODS: The Reading the Mind in the Eyes Test developed by Baron-Cohen and colleagues was used to assess the models' proficiency in interpreting visual emotional indicators. Simultaneously, the Levels of Emotional Awareness Scale was used to evaluate the large language models' aptitude in textual mentalizing. Collating data from both tests provided a holistic view of the mentalizing capabilities of ChatGPT-4 and Bard. RESULTS: ChatGPT-4, displaying a pronounced ability in emotion recognition, secured scores of 26 and 27 in 2 distinct evaluations, significantly deviating from a random response paradigm (P<.001). These scores align with established benchmarks from the broader human demographic. Notably, ChatGPT-4 exhibited consistent responses, with no discernible biases pertaining to the sex of the model or the nature of the emotion. In contrast, Google Bard's performance aligned with random response patterns, securing scores of 10 and 12 and rendering further detailed analysis redundant. In the domain of textual analysis, both ChatGPT and Bard surpassed established benchmarks from the general population, with their performances being remarkably congruent. CONCLUSIONS: ChatGPT-4 proved its efficacy in the domain of visual mentalizing, aligning closely with human performance standards. Although both models displayed commendable acumen in textual emotion interpretation, Bard's capabilities in visual emotion interpretation necessitate further scrutiny and potential refinement. This study stresses the criticality of ethical AI development for emotional recognition, highlighting the need for inclusive data, collaboration with patients and mental health experts, and stringent governmental oversight to ensure transparency and protect patient privacy.

The plasticity of ChatGPT's mentalizing abilities: personalization for personality structures.

This study evaluated the potential of ChatGPT, a large language model, to generate mentalizing-like abilities that are tailored to a specific personality structure and/or psychopathology. Mentalization is the ability to understand and interpret one's own and others' mental states, including thoughts, feelings, and intentions. Borderline Personality Disorder (BPD) and Schizoid Personality Disorder (SPD) are characterized by distinct patterns of emotional regulation. Individuals with BPD tend to experience intense and unstable emotions, while individuals with SPD tend to experience flattened or detached emotions. We used ChatGPT's free version 23.3 and assessed the extent to which its responses akin to emotional awareness (EA) were customized to the distinctive personality structure-character characterized by Borderline Personality Disorder (BPD) and Schizoid Personality Disorder (SPD), employing the Levels of Emotional Awareness Scale (LEAS). ChatGPT was able to accurately describe the emotional reactions of individuals with BPD as more intense, complex, and rich than those with SPD. This finding suggests that ChatGPT can generate mentalizing-like responses consistent with a range of psychopathologies in line with clinical and theoretical knowledge. However, the study also raises concerns regarding the potential for stigmas or biases related to mental diagnoses to impact the validity and usefulness of chatbot-based clinical interventions. We emphasize the need for the responsible development and deployment of chatbot-based interventions in mental health, which considers diverse theoretical frameworks.

ChatGPT outperforms humans in emotional awareness evaluations.

The artificial intelligence chatbot, ChatGPT, has gained widespread attention for its ability to perform natural language processing tasks and has the fastest-growing user base in history. Although ChatGPT has successfully generated theoretical information in multiple fields, its ability to identify and describe emotions is still unknown. Emotional awareness (EA), the ability to conceptualize one's own and others' emotions, is considered a transdiagnostic mechanism for psychopathology. This study utilized the Levels of Emotional Awareness Scale (LEAS) as an objective, performance-based test to analyze ChatGPT's responses to twenty scenarios and compared its EA performance with that of the general population norms, as reported by a previous study. A second examination was performed one month later to measure EA improvement over time. Finally, two independent licensed psychologists evaluated the fit-to-context of ChatGPT's EA responses. In the first examination, ChatGPT demonstrated significantly higher performance than the general population on all the LEAS scales (Z score = 2.84). In the second examination, ChatGPT's performance significantly improved, almost reaching the maximum possible LEAS score (Z score = 4.26). Its accuracy levels were also extremely high (9.7/10). The study demonstrated that ChatGPT can generate appropriate EA responses, and that its performance may improve significantly over time. The study has theoretical and clinical implications, as ChatGPT can be used as part of cognitive training for clinical populations with EA impairments. In addition, ChatGPT's EA-like abilities may facilitate psychiatric diagnosis and assessment and be used to enhance emotional language. Further research is warranted to better understand the potential benefits and risks of ChatGPT and refine it to promote mental health.