A Procedure for Classifying Attachments and Affective Social Bonds in Human-Robot Dyads

Imran Khan; Elisabeth Blagrove

cs.HCSep 24, 2026

A Procedure for Classifying Attachments and Affective Social Bonds in Human-Robot Dyads

Authors: Imran Khan, Elisabeth Blagrove

Organizations: Department of Computer Science, University of Warwick, Coventry, CV4 7AL, United Kingdom. · Department of Psychology, University of Warwick, Coventry, CV4 7AL, United Kingdom.

Abstract

Human-robot interaction (HRI) claims that people form attachments and social bonds with artificial agents, yet the terms are often applied without the behavioural and physiological criteria that give them content in their source disciplines. Without this empirical grounding, studies deploy widely divergent methods, frequently producing expansive relational claims that far outstrip their underlying evidence. To address this, we propose a standardised four-question procedure, grounded in criteria established in the developmental, ethological, and neuroendocrine literatures, that classifies a given human-robot tie as an attachment, an affective social bond, or no relationship, with intermediate classifications when evidence is incomplete. We specify minimum evidential requirements for each question, and provide candidate HRI study designs, adapted from validated human-human, human-animal, and animal-animal paradigms. We then demonstrate the procedure by applying it to a representative set of published HRI studies, showing how often relational claims outstrip what the reported designs can establish. Finally, we discuss the ethical and regulatory burdens created when artificial agents engage human biobehavioural systems. By replacing the divergent operationalisations with a unified, criterion-based classification, this paper gives HRI practitioners a standardised basis for evaluating, classifying, and comparing human-robot relationships, and sets out the experimental rigour that each classification demands. We therefore call on researchers of human-robot relationships to adopt such rigour, or to consider alternative terminology in their descriptions of these ties.

Figures & tables

Q1	Q2	Q3	Q4	Classification of the relationship
✓	✓	$\ast$	✓	Attachment. Both selectivity and security regulation requirements are met. Answer to Q3 not required.
✓	✓	✓	—	Affective social bond. Both comparative requirements met, with the channel profile reported separately.
✓	✓	—	—	Affiliative preference. Ainsworth’s affectional-bond criteria are met. However, regulation criterion not tested.
$\ast$	—	✓	—	Non-specific regulation. Physiological changes are measured with respect to robot’s presence, without evidence that the state change is dependant on the specific agent
✓	—	—	—	Affectional tie . Ainsworth’s affectional tie criteria met: Question 1 answered via observation. Selectivity of the agent and physiological regulation not tested.
—	$\cdot$	$\cdot$	$\cdot$	No relationship to classify. The affectional criteria are not met, usually because the encounter was too brief for persistence to be observable. The interaction may still be engaging, trusted and valued.

Table 1: The classification provided by each set of answers to the procedure presented in Section 4 , resulting in one of six relationship classification. ✓ established, — not established,

\ast

not required,

\cdot

not reached.

Regulator	Physiological substrate	What an agent supplies	Observable proxy	Reference method	Practical proxy
Phasic (Reactive/Event-Driven)
Stress buffering	HPA axis; cortisol response, CRF suppression ( Hostinar et al., 2014 )	Contingent responsiveness	Cortisol reactivity and recovery (HPA); EDA and HRV (autonomic)	Trier Social Stress Test ( Allen et al., 2016 ; Heinrichs et al., 2003 ) ; salivary cortisol AUC i ( Pruessner et al., 2003 )	Chest-strap ECG for HRV ( Schaffarczyk et al., 2022 ) ; wrist EDA ( Milstein and Gordon, 2020 )
Pain and threat modulation	Endogenous $\mu$ -opioid signalling ( Machin and Dunbar, 2011 ; Løseth et al., 2025 )	Presence; contact only if embodied	Pain threshold, tolerance and rating	Laser-evoked potentials and pain report ( von Mohr et al., 2018 ) ; threat-of-shock handholding ( Coan et al., 2017 )	Contact against no-contact under thermal pain ( Geva et al., 2020 ; Nakae et al., 2025 )
Tonic (Continuous/Baseline)
Behavioural and physiological synchrony	Interpersonal cardiorespiratory coupling ( Goldstein et al., 2017 ) ; motor and speech timing ( Ramseyer and Tschacher, 2011 ) ; raised pain thresholds under synchrony ( Cohen et al., 2010 )	Turn-taking, prosody and gesture mirroring, by fixed policy	Tonic autonomic state; movement, speech and cardiac cross-correlation	Motion Energy Analysis ( Ramseyer and Tschacher, 2011 ) ; cross-recurrence ( Coco and Dale, 2014 ) ; surrogate controls ( Palumbo et al., 2017 )	Fixed-camera video; chest-strap ECG for cardiac coupling ( Schaffarczyk et al., 2022 ) ; the agent’s own turn logs
Thermal and circadian co-regulation	Homeostatic regulation carried by contact ( Hofer, 1994 ; Fotopoulou et al., 2022 )	Requires contingent warmth and continuous presence	Skin temperature; sleep timing and continuity	Infrared thermography ( Nazzari et al., 2024 ; Filippini et al., 2020 )	Consumer actigraphy, for sleep timing and continuity ( Yuan et al., 2024 )

Table 2: The physiological regulators an affective social bond is composed of, grouped by when they operate.

Explore similar work

Sep 3, 2026cs.RO

Establishing a Dynamic Multimodal HRI Dataset for Engagement Analysis with a Humanoid Robot

This paper presents an experimental design for constructing a multimodal dataset to analyze user engagement in human-robot interaction (HRI). Prior studies have mainly relied on observable behavioral cues, with limited frameworks integrating physiological signals. We therefore propose a structured data-collection protocol to build a multimodal dataset that includes wearable physiological signals, behavioral data, and self-report measures under different levels of task complexity defined in this experiment.

Buwan Kim, Wonse Jo

Department of Information and Telecommunication Engineering, Incheon National University, Incheon, South Korea 22012.

Human-Robot Interaction Wearable Physiology Multimodal Dataset Analytics

Ask PDF

Jun 22, 2026cs.RO

When Robots Rate Their Own Interactions: Engagement Validity and the Strangeness Failure

Human-robot interaction (HRI) evaluation relies almost exclusively on human-completed questionnaires, leaving the robot's perspective unexamined. We propose an \textit{inverted evaluation}, in which LLM-powered robots complete the same standardized instruments from their own perspective, and test whether these ratings agree with human ground truth. In Study~~1, five LLMs completed HRI-CUES, Godspeed, and RoSAS questionnaires for 25~~interactions (

N = 1{,}522

evaluations) from the HRI-CUES dataset. LLMs achieved moderate-to-strong agreement on engagement dimensions (satisfaction

r

up to

.65

and enjoyment

r

up to

.72

) with excellent test-retest reliability (ICC

\geq .82

), but \textit{systematically inverted} the comfort/strangeness dimension (

r = -.44

-.67

, all

p < .05

), conflating engagement with comfort. In Study~~2, a Nao robot running Claude~~Sonnet~4.5 replicated these patterns in live interactions (

N = 4

), including real-time turn-by-turn assessment. The strangeness failure persisted across five models, synthetic controls, and embodied deployment for two participants. We argue that current LLM-based robots lack access to the internal affective states needed to assess constructs like strangeness, and that inverted evaluation requires supplementary modalities (e.g., physiological signals, gaze, proxemics) to move beyond behavioral proxies. These findings establish boundary conditions for using LLMs as interaction evaluators in HRI.

Victor Lockwood, Hasan Mahmud, Mohammad Javad Khojasteh +2

MS in AI Program Rochester Institute of Technology Rochester, New York, USA · School of Communication Rochester Institute of Technology Rochester, New York, USA · Electrical and Microelectronic Engineering Rochester Institute of Technology Rochester, New York, USA

Human-Ai Interaction Robot Systems Analytics Affective

Ask PDF

May 12, 2026cs.RO

Mapping Embodied Affective Touch Strategies on a Humanoid Robot

Affective touch in human-robot interaction is shaped not only by emotional intent, but also by robot embodiment, including touch location, physical constraints, and perceived agency or social role. Existing HRI studies typically focus on one or two isolated body parts, limiting understanding of how affective touch generalises across the full humanoid body. We present a study with 32 participants interacting with the iCub robot, which is equipped with full-body distributed tactile sensors. Participants expressed eight emotions under three conditions: free touch, arm-only touch, and torso-only touch. Results show that body region and spatial constraints jointly shaped both touch location and dynamics. In free touch, participants preferred socially accessible upper-body regions, while less frequently touched areas showed stronger emotion-specific selectivity. Emotion-related variation was more evident in motion features for arm-only touch and pressure features for torso-only touch. Touch strategies also did not transfer directly between free and constrained conditions, even within the same coarse body region. Participants reported increased closeness to the robot after interaction, with around 30 percent reporting a change in perceived social relationship. Together, these findings show that affective touch expression is strongly body-region dependent and shaped by embodiment constraints.

Qiaoqiao Ren, Omar Eldardeer, Francesca Cocchella +3

Faculty of Engineering and Architecture, 1IDLab-AIRO, Ghent University – imec, Technologiepark 126, 9052 Gent, Belgium · School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology · COgNiTive Architecture for Collaborative Technologies Unit of the Italian Institute of Technology

Human-Robot Interaction Embodiment Emotional Regulation

Ask PDF