Explaining GPT-4's Schema of Depression Using Machine Behavior Analysis

Use of large language models such as ChatGPT (GPT-4) for mental health support has grown rapidly, emerging as a promising route to assess and help people with mood disorders, like depression. However, we have a limited understanding of GPT-4's schema of mental disorders, that is, how it internally associates and interprets symptoms. In this work, we leveraged contemporary measurement theory to decode how GPT-4 interrelates depressive symptoms to inform both clinical utility and theoretical understanding. We found GPT-4's assessment of depression: (a) had high overall convergent validity (r = .71 with self-report on 955 samples, and r = .81 with experts judgments on 209 samples); (b) had moderately high internal consistency (symptom inter-correlates r = .23 to .78 ) that largely aligned with literature and self-report; except that GPT-4 (c) underemphasized suicidality's -- and overemphasized psychomotor's -- relationship with other symptoms, and (d) had symptom inference patterns that suggest nuanced hypotheses (e.g. sleep and fatigue are influenced by most other symptoms while feelings of worthlessness/guilt is mostly influenced by depressed mood).

SOCIALITE-LLAMA: An Instruction-Tuned Model for Social Scientific Tasks

Social science NLP tasks, such as emotion or humor detection, are required to capture the semantics along with the implicit pragmatics from text, often with limited amounts of training data. Instruction tuning has been shown to improve the many capabilities of large language models (LLMs) such as commonsense reasoning, reading comprehension, and computer programming. However, little is known about the effectiveness of instruction tuning on the social domain where implicit pragmatic cues are often needed to be captured. We explore the use of instruction tuning for social science NLP tasks and introduce Socialite-Llama -- an open-source, instruction-tuned Llama. On a suite of 20 social science tasks, Socialite-Llama improves upon the performance of Llama as well as matches or improves upon the performance of a state-of-the-art, multi-task finetuned model on a majority of them. Further, Socialite-Llama also leads to improvement on 5 out of 6 related social tasks as compared to Llama, suggesting instruction tuning can lead to generalized social understanding. All resources including our code, model and dataset can be found through bit.ly/socialitellama.

Comparing Human-Centered Language Modeling: Is it Better to Model Groups, Individual Traits, or Both?

Natural language processing has made progress in incorporating human context into its models, but whether it is more effective to use group-wise attributes (e.g., over-45-year-olds) or model individuals remains open. Group attributes are technically easier but coarse: not all 45-year-olds write the same way. In contrast, modeling individuals captures the complexity of each person's identity. It allows for a more personalized representation, but we may have to model an infinite number of users and require data that may be impossible to get. We compare modeling human context via group attributes, individual users, and combined approaches. Combining group and individual features significantly benefits user-level regression tasks like age estimation or personality assessment from a user's documents. Modeling individual users significantly improves the performance of single document-level classification tasks like stance and topic detection. We also find that individual-user modeling does well even without user's historical data.

Adaptive Language-based Mental Health Assessment with Item-Response Theory

Mental health issues widely vary across individuals - the manifestations of signs and symptoms can be fairly heterogeneous. Recently, language-based depression and anxiety assessments have shown promise for capturing this heterogeneous nature by evaluating a patient's own language, but such approaches require a large sample of words per person to be accurate. In this work, we introduce adaptive language-based assessment - the task of iteratively estimating an individual's psychological score based on limited language responses to questions that the model also decides to ask. To this end, we explore two statistical learning-based approaches for measurement/scoring: classical test theory (CTT) and item response theory (IRT). We find that using adaptive testing in general can significantly reduce the number of questions required to achieve high validity (r ~ 0.7) with standardized tests, bringing down from 11 total questions down to 3 for depression and 5 for anxiety. Given the combinatorial nature of the problem, we empirically evaluate multiple strategies for both the ordering and scoring objectives, introducing two new methods: a semi-supervised item response theory based method (ALIRT), and a supervised actor-critic based model. While both of the models achieve significant improvements over random and fixed orderings, we find ALIRT to be a scalable model that achieves the highest accuracy with lower numbers of questions (e.g. achieves Pearson r ~ 0.93 after only 3 questions versus asking all 11 questions). Overall, ALIRT allows prompting a reduced number of questions without compromising accuracy or overhead computational costs.

Systematic Evaluation of GPT-3 for Zero-Shot Personality Estimation

Very large language models (LLMs) perform extremely well on a spectrum of NLP tasks in a zero-shot setting. However, little is known about their performance on human-level NLP problems which rely on understanding psychological concepts, such as assessing personality traits. In this work, we investigate the zero-shot ability of GPT-3 to estimate the Big 5 personality traits from users' social media posts. Through a set of systematic experiments, we find that zero-shot GPT-3 performance is somewhat close to an existing pre-trained SotA for broad classification upon injecting knowledge about the trait in the prompts. However, when prompted to provide fine-grained classification, its performance drops to close to a simple most frequent class (MFC) baseline. We further analyze where GPT-3 performs better, as well as worse, than a pretrained lexical model, illustrating systematic errors that suggest ways to improve LLMs on human-level NLP tasks.

Human-Centered Metrics for Dialog System Evaluation

We present metrics for evaluating dialog systems through a psychologically-grounded "human" lens: conversational agents express a diversity of both states (short-term factors like emotions) and traits (longer-term factors like personality) just as people do. These interpretable metrics consist of five measures from established psychology constructs that can be applied both across dialogs and on turns within dialogs: emotional entropy, linguistic style and emotion matching, as well as agreeableness and empathy. We compare these human metrics against 6 state-of-the-art automatic metrics (e.g. BARTScore and BLEURT) on 7 standard dialog system data sets. We also introduce a novel data set, the Three Bot Dialog Evaluation Corpus, which consists of annotated conversations from ChatGPT, GPT-3, and BlenderBot. We demonstrate the proposed human metrics offer novel information, are uncorrelated with automatic metrics, and lead to increased accuracy beyond existing automatic metrics for predicting crowd-sourced dialog judgements. The interpretability and unique signal of our proposed human-centered framework make it a valuable tool for evaluating and improving dialog systems.

Transfer and Active Learning for Dissonance Detection: Addressing the Rare-Class Challenge

While transformer-based systems have enabled greater accuracies with fewer training examples, data acquisition obstacles still persist for rare-class tasks -- when the class label is very infrequent (e.g. < 5% of samples). Active learning has in general been proposed to alleviate such challenges, but choice of selection strategy, the criteria by which rare-class examples are chosen, has not been systematically evaluated. Further, transformers enable iterative transfer-learning approaches. We propose and investigate transfer- and active learning solutions to the rare class problem of dissonance detection through utilizing models trained on closely related tasks and the evaluation of acquisition strategies, including a proposed probability-of-rare-class (PRC) approach. We perform these experiments for a specific rare class problem: collecting language samples of cognitive dissonance from social media. We find that PRC is a simple and effective strategy to guide annotations and ultimately improve model accuracy while transfer-learning in a specific order can improve the cold-start performance of the learner but does not benefit iterations of active learning.

Robust language-based mental health assessments in time and space through social media

Compared to physical health, population mental health measurement in the U.S. is very coarse-grained. Currently, in the largest population surveys, such as those carried out by the Centers for Disease Control or Gallup, mental health is only broadly captured through "mentally unhealthy days" or "sadness", and limited to relatively infrequent state or metropolitan estimates. Through the large scale analysis of social media data, robust estimation of population mental health is feasible at much higher resolutions, up to weekly estimates for counties. In the present work, we validate a pipeline that uses a sample of 1.2 billion Tweets from 2 million geo-located users to estimate mental health changes for the two leading mental health conditions, depression and anxiety. We find moderate to large associations between the language-based mental health assessments and survey scores from Gallup for multiple levels of granularity, down to the county-week (fixed effects $\beta = .25$ to $1.58$; $p<.001$). Language-based assessment allows for the cost-effective and scalable monitoring of population mental health at weekly time scales. Such spatially fine-grained time series are well suited to monitor effects of societal events and policies as well as enable quasi-experimental study designs in population health and other disciplines. Beyond mental health in the U.S., this method generalizes to a broad set of psychological outcomes and allows for community measurement in under-resourced settings where no traditional survey measures - but social media data - are available.

Human Language Modeling

Natural language is generated by people, yet traditional language modeling views words or documents as if generated independently. Here, we propose human language modeling (HuLM), a hierarchical extension to the language modeling problem whereby a human-level exists to connect sequences of documents (e.g. social media messages) and capture the notion that human language is moderated by changing human states. We introduce, HaRT, a large-scale transformer model for the HuLM task, pre-trained on approximately 100,000 social media users, and demonstrate its effectiveness in terms of both language modeling (perplexity) for social media and fine-tuning for 4 downstream tasks spanning document- and user-levels: stance detection, sentiment classification, age estimation, and personality assessment. Results on all tasks meet or surpass the current state-of-the-art.

Understanding RoBERTa's Mood: The Role of Contextual-Embeddings as User-Representations for Depression Prediction

Many works in natural language processing have shown connections between a person's personal discourse and their personality, demographics, and mental health states. However, many of the machine learning models that predict such human traits have yet to fully consider the role of pre-trained language models and contextual embeddings. Using a person's degree of depression as a case study, we do an empirical analysis on which off-the-shelf language model, individual layers, and combinations of layers seem most promising when applied to human-level NLP tasks. Notably, despite the standard in past work of suggesting use of either the second-to-last or the last 4 layers, we find layer 19 (sixth-to last) is the most ideal by itself, while when using multiple layers, distributing them across the second half(i.e. Layers 12+) of the 24 layers is best.