The Psychology of Gen AI Series

The Psychology of Gen AI: Beyond Review: Why Human Judgment Remains Essential in AI-Assisted Research Design

As generative AI becomes increasingly integrated into research workflows, what does “human in the loop” actually mean? This latest installment in the ARF and MSI’s Psychology of Gen AI series explores how researchers and large language models can work together to develop rigorous research designs—and why human judgment remains indispensable throughout the process. By comparing multiple, AI-generated experimental designs and refining them through expert evaluation, the study demonstrates that AI is most valuable as a collaborator that expands possibilities, not as a replacement for methodological expertise.

The Psychology of Gen AI: Beyond the Recommendation: How Prompt Language Changes How AI Frames Your Brand

Generative AI tools are not just recommending products—they are shaping how consumers perceive them. This latest issue in the Psychology of Gen AI series, the third phase of the seventh study on AI product recommendations, examines how small changes in prompt wording alter the explanations AI systems generate around the same brand. Using ChatGPT descriptions of Arm & Hammer Advance White toothpaste across multiple, shopping-related prompts, the study reveals how AI recommendations construct different product narratives, emphasize different benefits and even introduce different caveats depending on the consumer’s wording of their question.

The Psychology of Gen AI: Finding Your Brand’s AI Niche: How Prompt Nuance Shapes Product Recommendations

As generative AI becomes a key part of how consumers discover and evaluate products, a new question emerges for marketers: how can they ensure their brands show up in AI-driven recommendations? This ARF and MSI experiment, the second phase of the seventh study in the Psychology of Gen AI series, reveals that even small changes in prompt wording can significantly influence which brands appear—helping non-market dominant brands carve out visibility by aligning with specific product attributes rather than competing broadly for “best” status.

What AI Recommends—and Why: Inside the Logic of “Best” Product Choices

As generative AI tools increasingly shape how consumers search, shop, compare and evaluate products, understanding how they make recommendations has become critical for marketers. This seventh experiment in our ongoing Psychology of Gen AI series, is the first phase in a study that examines how large language models (LLMs), like ChatGPT and Claude, determine what qualifies as the “best” product—and reveals that their recommendations are far from neutral. Instead, they tend to rely on narrow, repetitive sets of familiar brands and structured response patterns that may reinforce existing market leaders. The findings highlight important implications for brand visibility, competitive dynamics and how marketers should position their products in AI-driven environments.

When Language Becomes Targeting: How Gender Cues Shape AI Recommendations

As generative AI tools increasingly influence product discovery and decision-making, subtle cues in user language can shape what consumers are shown—and how options are framed. This research examines how implicit and explicit gender signals affect AI-generated product recommendations, revealing systematic differences in categories, brand repetition, descriptive language and price information. The findings raise important questions for advertisers and researchers about bias, brand visibility and the growing cultural role of AI in shaping consumer norms.

When Style Becomes Signal: How Gendered Language Shapes Generative AI Output

As generative AI tools become embedded in advertising and marketing research workflows, questions about bias increasingly extend beyond outputs to the interaction itself. This study examines whether gendered patterns can enter AI through subtle differences in how prompts are phrased. By systematically varying linguistic styles using psychologically grounded traits, the research shows that implicit, style-based, gender cues shape AI prompt construction more strongly than explicit, gender labels, with important implications for how bias may propagate upstream in AI-assisted marketing and research applications.

Why Synthetic Respondents Flatten Consumer Sentiment

A new ARF Psych of GenAI experiment reveals that large language models apply a rigid, rule-driven logic when evaluating privacy scenarios—even when humans typically shift their reasoning based on framing, emotion and social context. Unlike consumers, who blend intuition, feeling and social perspective into their judgments, GPT-4o relied on a single internal rule across all testing conditions: data use is acceptable only with explicit consent. This consistency offers value for certain analytic tasks but exposes limits for advertising research that depends on emotional nuance and context-sensitive consumer insight.

Steering AI Bias: How Persona Prompts Unlock Nuance in Gen AI Responses

Large language models mirror human cognitive biases—but can those biases be guided? New ARF and MSI research reveals that while loss aversion remains deeply ingrained in AI responses, introducing persona information, such as demographics or personality traits, can increase variability and make outputs more nuanced. For advertisers and researchers, this opens the door to design strategic prompts that spark richer and more nuanced, human-like responses.

The Bias Beneath the Average: What Loss Aversion Reveals About How AI Thinks

This experiment reveals how models like ChatGPT not only replicate human cognitive biases, such as loss aversion, but also compress variability into uniform patterns. This raises concerns for advertising researchers who rely on authentic insights into consumer behavior.

Acting the Part: Can AI Think Like a CFO? Using Personas to Test Generative AI’s Strategic Reasoning

The ARF tested whether generative AI can adopt executive personas and provide credible, role-specific strategies. This experiment highlights how AI performs when “thinking like” organizational leaders, its limitations in institutional logic and feasibility, and how human-in-the-loop feedback can refine outputs and create nuanced and worthwhile results.

PG VS R: The Psychology of Prompted Thought

Can sanitized AI tools truly capture the nuance required for advertising and brand research? Is a less restrained one more likely to produce skewed results? This comparative deep dive, from ARF and MSI explores how two popular large language models—ChatGPT-4o and Grok 3—respond when prompted with complex topics. The findings highlight how content moderation affects not only tone and specificity, but the very boundaries of inquiry. For advertising researchers navigating sensitive brand perception topics, understanding these model tradeoffs is essential.

Alternative Explanations: Can AI Rethink Its Own Reasoning?

Can AI challenge its own conclusions rather than merely reinforcing them? In this ARF experiment, researchers explored whether large language models (LLMs) like ChatGPT can go beyond efficiency and exhibit deeper critical thinking skills. By prompting AI to evaluate and compare hypotheses—including its own—this study reveals how LLMs can serve as interpretive collaborators in research and theoretical reasoning.