SYNTHESIS NOTE
Topics›Psychology Therapy Practice›this note

Can we control personality in language models without prompting?

Can lightweight adapter modules enable continuous, fine-grained control over psychological traits in transformer outputs independent of prompt engineering? This explores whether architecture-level personality modification outperforms prompt-based approaches.

Synthesis note · 2026-02-23 · sourced from Psychology Therapy Practice

PsychAdapter modifies the transformer architecture to accept continuous psychological trait scores as input, enabling generation conditioned on personality, mental health, and demographic variables without consuming context window or relying on prompt engineering. The key difference from prior work: trait influence is applied at every transformer layer via a learned dimension expansion, not just at the input level.

Training uses social media and blog posts with estimated psychological scores from an empirically-trained language-based assessment model. The adapter learns how to weight the psychological scores' contribution to each layer alongside standard next-word prediction. The result: fine-grained, continuous control over personality expression. An input vector of (0, 0, +3, 0, 0) generates text characteristic of high extraversion while remaining average on other Big Five dimensions. Any combination is possible, including interactions: high openness with low extraversion produces text that captures both traits simultaneously.

Expert raters evaluated generated text at 87.3% average accuracy for Big Five personalities and 96.7% for depression and life satisfaction. These numbers hold across GPT-2, Gemma (2B), and Llama 3, demonstrating model-agnostic applicability. The total added parameters are less than 0.1% of the base model (55,296 for Gemma 2B vs 2 billion base parameters), making distribution trivial.

Applications extend beyond mental health simulation: customer service training with diverse personalities, crisis worker training with simulated distress levels, machine translation matched to audience education/dialect levels, and research tools that generate coherent text (not isolated words/phrases) for trait analysis. Since Do personality traits activate hidden emoji patterns in language models?, PsychAdapter may be activating these pre-existing trait-language circuits through a more precise mechanism than prompting or full fine-tuning.

Inquiring lines that read this note 62

This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.

Can persona prompting reliably change how language models behave? Does preference-optimized training sacrifice accuracy and calibration for perceived quality? How can persona-based systems overcome their inherent limitations in modeling diversity? What factors determine AI persuasion effectiveness and what controls mitigate it? Does prompting inject new knowledge into models or activate existing knowledge? What distinguishes emotionally supportive AI from AI that fosters unhealthy dependence? What determines when personalization helps versus harms model behavior? What drives sycophancy in language models and can it be fixed? How do prompt design choices and framing measurably influence model outputs? How does model scale affect fine-tuning and downstream task performance? What constraints limit conversational AI's effectiveness in therapeutic contexts? Do engagement-optimized chatbots create dependencies that undermine user autonomy? Does learning articulatory physics explain speech communication superiority? How can AI systems maintain consistent personas across conversations? What training data structures and signals produce robust learning across distributions? What mechanisms cause models to behave deceptively despite alignment training? What distinguishes genuine understanding from correct linguistic performance? What can prompting techniques actually accomplish in language models? Can synthetic data preserve necessary diversity to prevent model collapse? Do language models learn underlying linguistic structure or merely surface patterns? How can we distinguish genuine model self-knowledge from behavioral performance? How do transformer circuits and attention mechanisms enable or limit reasoning? Do language models genuinely understand or primarily match statistical patterns? Do language models truly understand their encoded knowledge or merely retrieve patterns? Should GUI agents use structured UI perception instead of raw pixels?

Related concepts in this collection 3

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
13 direct connections · 98 in 2-hop network ·medium cluster Open in graph ↗
Can we control personality in language models wi… Do personality traits activate hidden emoji patter… Can we track and steer personality shifts during m… Can open language models adopt different personali…

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Search by related questions 5

Suggested questions this note speaks to — click to search the collection, or type your own.

Do personality traits occupy specific mechanistic locations in pretrained models?Can continuous persona vectors in activation space monitor personality shifts?Why do most open language models resist personality conditioning via prompts?How do trait adapters interact with different base model architectures?Can personality control improve training outcomes for crisis workers and therapists?
Original note title

lightweight psychological trait adapters modify every transformer layer with less than 0.1 percent additional parameters — enabling fine-grained psychological profile control independent of prompting