Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Beyond Input-Output: Rethinking Creativity through Design-by-Analogy in Human-AI Collaboration

Xuechen Li, Shuai Zhang, Nan Cao, Qing Chen

TL;DR

Foundation models risks homogenizing creativity; this work conducts a PRISMA-based systematic review of Design-by-Analogy (DbA) to map representations and a four-phase workflow across three domains. It reframes DbA as a technological mediator that augments human creativity by structuring cross-domain analogy through vision, inspiration, ideation, fabrication, evaluation, and meta stages. Key contributions include a six-representation taxonomy, a four-phase DbA workflow with seven stages, and guidelines addressing user, data, algorithms, and interaction, plus domain-specific applications and ethical considerations. The findings support DbA as a versatile, human-centered creativity-support layer for human-AI collaboration and outline concrete directions for responsible future design research.

Abstract

While the proliferation of foundation models has significantly boosted individual productivity, it also introduces a potential challenge: the homogenization of creative content. In response, we revisit Design-by-Analogy (DbA), a cognitively grounded approach that fosters novel solutions by mapping inspiration across domains. However, prevailing perspectives often restrict DbA to early ideation or specific data modalities, while reducing AI-driven design to simplified input-output pipelines. Such conceptual limitations inadvertently foster widespread design fixation. To address this, we expand the understanding of DbA by embedding it into the entire creative process, thereby demonstrating its capacity to mitigate such fixation. Through a systematic review of 85 studies, we identify six forms of representation and classify techniques across seven stages of the creative process. We further discuss three major application domains: creative industries, intelligent manufacturing, and education and services, demonstrating DbA's practical relevance. Building on this synthesis, we frame DbA as a mediating technology for human-AI collaboration and outline the potential opportunities and inherent risks for advancing creativity support in HCI and design research.

Beyond Input-Output: Rethinking Creativity through Design-by-Analogy in Human-AI Collaboration

TL;DR

Foundation models risks homogenizing creativity; this work conducts a PRISMA-based systematic review of Design-by-Analogy (DbA) to map representations and a four-phase workflow across three domains. It reframes DbA as a technological mediator that augments human creativity by structuring cross-domain analogy through vision, inspiration, ideation, fabrication, evaluation, and meta stages. Key contributions include a six-representation taxonomy, a four-phase DbA workflow with seven stages, and guidelines addressing user, data, algorithms, and interaction, plus domain-specific applications and ethical considerations. The findings support DbA as a versatile, human-centered creativity-support layer for human-AI collaboration and outline concrete directions for responsible future design research.

Abstract

While the proliferation of foundation models has significantly boosted individual productivity, it also introduces a potential challenge: the homogenization of creative content. In response, we revisit Design-by-Analogy (DbA), a cognitively grounded approach that fosters novel solutions by mapping inspiration across domains. However, prevailing perspectives often restrict DbA to early ideation or specific data modalities, while reducing AI-driven design to simplified input-output pipelines. Such conceptual limitations inadvertently foster widespread design fixation. To address this, we expand the understanding of DbA by embedding it into the entire creative process, thereby demonstrating its capacity to mitigate such fixation. Through a systematic review of 85 studies, we identify six forms of representation and classify techniques across seven stages of the creative process. We further discuss three major application domains: creative industries, intelligent manufacturing, and education and services, demonstrating DbA's practical relevance. Building on this synthesis, we frame DbA as a mediating technology for human-AI collaboration and outline the potential opportunities and inherent risks for advancing creativity support in HCI and design research.
Paper Structure (25 sections, 8 figures, 1 table)

This paper contains 25 sections, 8 figures, 1 table.

Table of Contents

  1. Introduction
  2. Background
  3. Design-by-Analogy in Creativity Support
  4. Related Surveys
  5. Methodology
  6. Data Collection
  7. PRISMA Selection
  8. Validation and Corpus Overview
  9. Representations in Design-by-Analogy
  10. Design-by-Analogy Techniques in the Creative Process
  11. Phase 1: Product Definition
  12. Phase 2: Product Ideation
  13. Phase 3: Product Implementation
  14. Phase 4: Product Evaluation
  15. Applications
  16. ...and 10 more sections

Figures (8)

  • Figure 1: Overview of the PRISMA selection process used to construct the final corpus.
  • Figure 2: Statistical overview of the 85 paper corpus. The pie chart on the left illustrates the distribution by research focus, with application-oriented studies (48.2%) forming the largest category. The stacked bar chart on the right displays the temporal distribution of publications from 2005 to 2025, with colors representing the breakdown by venue categories, such as Human-Computer Interaction and Artificial Intelligence. Detailed corpus information can be found in the Appendix.
  • Figure 3: A taxonomy of six representations for Design-by-Analogy: (1) Semantics & Text jiayang2023storyanalogy, (2) Visual & Appearance warner2023interactive, (3) Material & Structure fischer2024nerf, (4) Function & Attribute fu2014bio, (5) Interaction & Experience kim2023star, and (6) Unconventional Contexts gentner1983structurelinsey2008modality.
  • Figure 4: From the opening image (Phase 1: Vision) to the final illustration (Phase 4: Meta), we use metaphor and narrative to depict the mediation role of the DbA in create process. Imagine a person standing before a wall. The DbA acts as a foundation of knowledge, elevating their perspective and enabling them to gain a bigger vision and inspiration. They then select content that aligns with their intrinsic value in order to generate ideas and begin prototyping. The DbA system then provides relevant information and details, assists with completing multiple specific operations, and subsequently transfers knowledge for usability evaluation. See Table \ref{['tab:Literature_Distribution']} for the classification of related papers corresponding to each stage.
  • Figure 5: Part of systematic review result of Design-by-Analogy (DbA). The table classifies 41 surveyed systems and studies along five key dimensions: (1) the supported Computational Stages stages gentner2011computationaljiang2022data, (2) the targeted Create Process stages, (3) the Representation types used, (4) the Level of Automation, and (5) the Specific Application Domain. The legend at the bottom provides a detailed explanation of all abbreviations, symbols, and color-coding.
  • ...and 3 more figures