Search

In this study, we examine whether we can replicate the findings of Stevenson and colleagues (2022), where the creative ability of GPT-3 was compared to human creativity on the Alternative Uses Test (AUT) - one of the most widely used verbal creativity tests – to establish the current state of AI's ability to come up with original and useful ideas. However, developments in this field are rapid and plenty – making the original study possibly already outdated. Moreover, as the authors mentioned, the comparison was not quite fair, as there was a difference in language (Dutch responses translated to English for humans and English responses for GPT-3) and there were differences in the instructions for GPT-3 and humans, where standard instructions were used for humans and optimized instructions for GPT-3. With this study, we want to replicate the findings of Stevenson et al. (2022) with minimal task differences for human participants and GPT runs, we will use the newest GPT engine (GPT-3.5-turbo), and we want to use a larger sample size. We will use the same scoring methods as in the previous study, as well as new ones to improve upon the flexibility measure, a measure that is often confounded with fluency. Also, in order to assess the response sets as a whole instead of assessing each response individually we will include snapshot creativity (Silvia et al., 2009). With this study, we not only want to compare GPT's and people's creative abilities, but also gain insight into the creative process of both and bring to light any similarities and differences between them.

Systematic review and a meta-analysis of the effects of priming on creativity: Protocol for data analysis. See attachment for details.

The objective of this study is to examine the influence of instructions (focused on creativity vs. personal relevance) and problem scope (narrow vs. broad problem) on problem construction performance. We look at how the problem construction performance subsequently influences performance during solution generation and selection.

Research has been struggling to establish a connection between mental illness and creativity due to high variability in conceptualizing and measuring both creativity and mental disorders. In this project, we want to untangle some of this confusion by focusing on positive schizotypy, a spectrum of symptoms that have been connected to creativity both on a clinical and non-clinical level and examining its connection to both creative performance behaviors (CPBs) – idea generation and exploration - and creative outcomes. Due to reduced latent inhibition in positive schizotypy, we propose improved divergent thinking while convergent thinking is impaired. This might lead to an imbalance of CPBs towards idea generation. Furthermore, reduced ability to shift between both thinking styles could impair the translation from CPBs into creative outcomes when positive schizotypy is high. These hypotheses are examined in a two-study design, measuring positive schizotypy and creativity with various self-report and task-based measurements.

Pre-registration for the A-AUT (Amsterdam Alternative Uses Task) validation study

In this study we want to examine how creative thinking influences negotiations. We expect that divergent thinking will positively influence the joint outcomes in an integrative negotiation. Additionally, we will examine whether participant’s (high) epistemic motivation weakens the positive effect of divergent thinking on negotiations.

Replication study

Research has robustly shown that diversity enhances creativity. However, one largely neglected question is whether such relationship can be reciprocal. Namely, does creativity inspire a preference for diversity? In the proposed research, we will examine whether creative thinking can improve people’s attitudes towards ethnically diverse individuals using explicit (Study 1), implicit (Study 2) and behavioral measures (Study 3). Building on research showing that creativity promotes cognitive, motivational and emotional changes, we propose that the mere act of thinking creatively has the potential to counteract mechanisms often responsible for a dislike and distancing from diverse individuals.

Previous research on ADHD and ASD has mainly focused on the deficits associated with these conditions, but there is also evidence for strengths. Unfortunately, our understanding of potential strengths in neurodevelopmental conditions is limited. One particular strength, creativity, has been associated with both ADHD and ASD. However, the distinct presentations of both conditions beg the question whether ADHD and ASD associate with the same or different aspects of creativity. Therefore, the current study investigated the links between ADHD and ASD symptoms, creative thinking abilities, and creative achievements. To investigate the spectrum of ADHD and ASD symptoms, self-reported ADHD and ASD symptoms, convergent (Remote Associations Test) and divergent thinking (Alternative Uses Task) and creative achievements (Creative Achievement Questionnaire) were assessed in a self-reportedly healthy sample of adults (n = 470). We performed correlation analysis to investigate the relation between ADHD/ASD symptoms and creativity measures. In a second phase of analysis, data from an adult ADHD case-control study (n = 151) were added to investigate the association between ADHD symptoms and divergent thinking in individuals with and without a diagnosis of ADHD. Our analysis revealed that having more ADHD symptoms in the general population was associated with higher scores on all the outcome measures for divergent thinking (fluency, flexibility, and originality), but not for convergent thinking. Individuals with an ADHD diagnosis in the case-control sample also scored higher on measures of divergent thinking. Combining data of the population based and case-control studies showed that ADHD symptoms predict divergent thinking up to a certain level of symptoms. No significant associations were found between the total number of ASD symptoms and any of the creativity measures. However, explorative analyses showed interesting links between the ASD subdomains of problems with imagination and symptoms that relate to social difficulties. Our findings showed a link between ADHD symptoms and divergent thinking abilities that plateaus in the clinical spectrum of symptoms. For ASD symptoms, no relation was found with creativity measures. Increasing the knowledge about positive phenotypes associated with neurodevelopmental conditions and their symptom dimensions might aid psychoeducation, decrease stigmatization and improve quality of life of individuals living with such conditions.

The evaluation of creative ideas is a crucial step in the innovation process. However, doing so accurately is a challenge; anecdotal and research evidence suggests that creative potential often gets lost during evaluation and selection. In this chapter, we review the literature on idea evaluation and discuss various models that have been proposed. Based on our review, we conclude that effective evaluation of creative ideas depends on two factors: (1) openness: evaluators need to be willing and able to consider and appreciate novel ideas, despite the uncertainty these ideas bring; and (2) expertise: evaluators need sufficient domain knowledge to assess whether an idea is (or can be) a valuable contribution. Finally, we discuss implications for research and practice and identify potential avenues for future research.