HCI Papers Cite HCI Papers, Increasingly So (2303.07539v2)

Abstract: To measure how HCI papers are cited across disciplinary boundaries, we collected a citation dataset of CHI, UIST, and CSCW papers published between 2010 and 2020. Our analysis indicates that HCI papers have been more and more likely to be cited by HCI papers rather than by non-HCI papers.

• The paper highlights a significant increase in internal citations among HCI papers over the past decade.
• It employs DOI data and tools like the Citation Chaser app to analyze citation patterns across major HCI venues.
• Findings indicate that while older HCI works retain broader external influence, recent papers are cited less by non-HCI fields.

Interdisciplinary Citation Trends in Human-Computer Interaction Research

In "HCI Papers Cite HCI Papers, Increasingly So," Xiang ‘Anthony’ Chen presents a citation analysis of Human-Computer Interaction (HCI) papers, aiming to examine the interdisciplinary nature and citation trends within and beyond the HCI domain. The paper focuses on HCI papers from notable venues such as CHI, UIST, and CSCW, published between 2010 and 2020.

Methodology

The paper undertakes a comprehensive citation analysis using DOI data from the CHI, UIST, and CSCW conference papers as sourced from the "What The HCI" website. By employing the Citation Chaser app and Lens.org API, the research compiles citation data covering these HCI papers and categorizes them into citations from core HCI versus non-HCI venues. A crucial step in the methodology is defining the core HCI venues list, derived primarily from SIGCHI-sponsored conferences and Google's "Human Computer Interaction" publication category. The research applies keyword-matching to determine citation sources, highlighting interdisciplinary boundaries.

Key Findings

The analysis reveals several noteworthy trends in the citation dynamics of HCI papers:

1. Internal Citation Increase: Over the examined decade, there is a notable trend where HCI papers have been increasingly cited by other HCI papers rather than by papers from outside the core HCI venues. This indicates a strengthening internal citation culture within the HCI discipline.
2. Temporal Influence on Citation Patterns: Figures in the paper demonstrate that recent HCI papers are cited less frequently by non-HCI venues when compared to earlier HCI papers. Even when controlling for an equal post-publication window of five years, this trend persists, suggesting a narrowing influence of HCI research beyond its traditional boundaries over time.
3. Dominance of Older Papers in Cross-Disciplinary Citations: Within any given citation year, older HCI papers tend to be cited more by non-HCI venues than newly published ones. This indicates a lasting visibility and influence of foundational HCI work outside its immediate field.
4. Recent Decline in External Citations: The research also finds a recent decline in the likelihood of HCI papers being cited by non-HCI papers, further suggesting a potential inward shift in the discipline's citation patterns.

Discussion and Implications

These findings carry significant implications for understanding the evolving nature of interdisciplinary research within HCI. The increasing trend of internal citations may reflect a consolidation phase within the field, where the focus is on refining and building upon existing bodywork. It raises critical questions about the extent of HCI's engagement with and impact on other fields, potentially attributing the observed decline to a diversification of topics or specialization within HCI venues themselves.

The limitations specified, such as potential bias due to incomplete venue lists and reliance on the Lens.org API, are essential considerations for interpreting the dataset's robustness and accuracy. Future studies could expand this analysis by incorporating more sophisticated classification techniques and additional data sources to acquire a broader, interdisciplinary citation landscape.

Future Research Directions

This research invites further exploration into various dimensions of citation dynamics across academic disciplines. Analyzing other fields, like Computer Vision or Artificial Intelligence, might provide comparative insights into citation practices and interdisciplinary impacts. Moreover, investigating the underlying causes of the decreasing external citations trend could offer actionable insights for researchers aiming to enhance the influence of their work beyond their immediate academic communities.

In conclusion, while the analyzed HCI papers indicate a prevalent trend of internal citation, which may speak to the maturity and autonomy of the field, they also call for a re-evaluation of engagement with external disciplines crucial for fostering broader interdisciplinary innovation.