Research Group: Accommodation and Social Categorization

The Accommodation and Social Categorization Research Group focused on the phenomenon of accommodation, where speakers adjust their verbal and non-verbal behavior in interactions, either by converging (becoming more similar) or diverging (accentuating differences) from their conversation partners. This phenomenon, also known by terms like convergence, alignment, and synchronization, occurs in speech, written communication, and even human-machine interactions.

Research Focus:

Vocal and Linguistic Accommodation:

Accommodation was studied across different contexts, including dialect contact situations, developmental aspects, human-computer interactions, and agreement/code-mixing in bilingual and multilingual environments. The group examined accommodation at various levels—lexicon, grammar, pronunciation, and non-verbal cues such as facial expressions and body movements.

Mechanisms Behind Accommodation:

Despite extensive research, it remained controversial whether accommodation resulted from low-level automatic mechanisms (linking perception and production) or more complex factors like mutual liking, phonetic distance, or conversational dynamics. The group explored this by analyzing the linguistic and phonetic convergence across a range of tasks and contexts, comparing instances of accommodation in both human-human and human-machine interactions.

Multidisciplinary Approach:

The group combined linguistic, acoustic, articulatory, and perceptual data to investigate the mechanisms and functions behind linguistic and phonetic convergence. They aimed to understand how accommodation contributes to language variation and change, and its implications for human and machine speaker identification, especially in forensic settings.

Key Findings:

• Accommodation was observed not only in spontaneous speech but also in controlled tasks, across different levels of communication.
• The group contributed to ongoing debates about the nature of accommodation, whether it is driven by automatic processes or influenced by social and contextual factors.
• Research extended to how accommodation impacts language variation, change, and the field of speaker recognition, with applications in forensic phonetics.

Collaborative Networks:

The research group collaborated with an international and multidisciplinary network, including experts in human-machine interactions (Bern Möbius, Gabriel Skantze), human speech communication (Jennifer Pardo), and forensic phonetics (Thayabaran Kathiresan, Andrea Fröhlich, Anil Alexander), advancing understanding in both theoretical and applied domains of vocal accommodation.

Through these efforts, the group contributed to a deeper understanding of the dynamic and multifaceted nature of linguistic and vocal accommodation, influencing studies on language evolution, machine learning, and forensic applications.