DALL-E interprets Hamlet’s personality using a stanine colour coded GPT4 prompt
It’s now February 2024, and time for another attempt at obtaining DALL-E images that represent literary characters. Forgetting about Great Aunt Georgiana for a moment, let’s revert to the more general questions:
- Can GPT4 create a Psychological Profile of a person
- Can it also work effectively with psychometric stanine scales?
- If so, what traits can it assign stanine values to?
- Can it assign colours to each trait, based on reasonable grounds?
- Can it prompt DALL-E to create images that represent psychological depictions of these characters?
GPT4 assures me that it can create a Psychological Profile that represent the extent of each trait on stanine scales in an abstract artistic representation and sees this as a creative way to visualize psychological traits or character attributes. It says that
“Each stanine value could be associated with specific visual cues such as:
- Color Intensity: The saturation or brightness of a color could represent the stanine score, with brighter or more saturated colors indicating higher scores.
- Color Quantity: The amount of a specific color used in the image could correlate with its stanine score, with more extensive use indicating a higher score.
- Shape Size: If shapes are used alongside or in place of colors, the size of these shapes could represent the stanine score, with larger shapes indicating higher scores.
- Placement or Frequency: The placement or frequency of elements within the composition could also reflect their importance or score on the stanine scale.”
However, when prompted for a psychological profile for Hamlet, all the stanine trait scores were positive (that is, above the estimated population average for norm referenced test results). A further prompt led it to identify several traits on which Hamlet’s scores might be less than average.
Its final list of Hamlet’s stanine scores and associated colours were:
April 2025 Update
It’s now 29th April 2025 and time for an update on this task using GPT-4o-mini
GPT4: Here is Hamlet rendered as a horizontal watercolor portrait in Georgiana Houghton’s abstract spirit-drawing style. The piece layers her signature palette—gamboge yellow halo for introspection and moral clarity, ultramarine and cobalt blues for intellectual depth, crimson lake accents for passion and love, emerald green highlights for self-control, and white filigree spirals for spiritual insight—woven together to reflect Hamlet’s introspective, morally complex, and emotionally rich character. So much better!
December 2025 Update
December 2025 Update
By December 2025, this experiment had developed into a more abstract Chromia-style approach. Instead of asking for “Hamlet” directly, the prompt was recast as a psycho-ethical abstraction:
Persistent recursive self-monitoring under conditions of moral uncertainty.
Heightened ethical sensitivity generating inhibition rather than clarity.
Tension between perceived obligation to act and fear of irreversible moral error.
Oscillation between decisiveness and withdrawal without sustained resolution.
Cognitive over-analysis functioning as both insight and defence against action.
Sustained internal conflict producing psychological stasis rather than collapse.
The resulting image marked a transition from character illustration toward the later Chromia/Chromos method of abstract psychological portraiture.
The experiment began as an attempt to see whether GPT-4 and DALL-E could translate a literary character into a psychometric-style visual abstraction. By 2025, the same line of work had begun to shift toward the later Chromia/Chromos method: not illustration of a named character, but abstract translation of psychological and ethical dynamics.
In retrospect, this Hamlet experiment was one of the early bridges between psychometric profiling, literary psychology, AI image generation, and the later Persona 2026 system.
This marked a transition from illustrating Hamlet as a named literary character toward the later Chromia/Chromos method: abstract psychological portraiture based on dynamics rather than depiction.