Organizations: 1,2,4Department of Mechanical Engineering, Indian Institute of Technology Kanpur, Kanpur, 208016, U.P, India. · Department of Humanities and Social Sciences, Indian Institute of Technology Kanpur, Kanpur, 208016, U.P, India.
Responding to one's name is among the earliest-emerging social orienting behaviors and is one of the most prominent aspects in the detection of Autism Spectrum Disorder (ASD). Typically developing children exhibit near-reflexive orienting to their name, whereas children with ASD often demonstrate reduced frequency, increased latency, or atypical patterns of response. In this study, we examine differential responsiveness to quantify name-calling stimuli delivered by both human agents and NAO, a humanoid robot widely employed in socially assistive interventions for autism. The analysis focuses on multiple behavioral parameters, including eye contact, response latency, head and facial orientation shifts, and duration of sustained interest. Video-based computational methods were employed, incorporating face detection, eye region tracking, and spatio-temporal facial analysis, to obtain fine-grained measures of children's responses. By comparing neurotypical and neuroatypical groups under controlled human-robot conditions, this work aims to understand how the source and modality of social cues affect attentional dynamics in name-calling contexts. The findings advance both the theoretical understanding of social orienting deficits in autism and the applied development of robot-assisted assessment tools.
Figures & tables
Figure 1: The Experimental setup
Parameter
Value
Subjects (Total)
75 children (45 TD, 30 ASD)
Manual data used
63 children (39 TD, 24 ASD)
Videos Used
116 (31*3 TD + 23 ASD)
Duration per Session
5-7 minutes per child
Name-Calling Instances
15 (3 per stimulus, randomized order)
Stimuli Types
5 (SM, SW, NM, NW, NR)
Table 1: Data Specifications and Recording Details
Stimulus
Mean EOP (%)
Duration (s)
Latency (s)
SM
68.42 ± 9.14
5.21 ± 1.37
1.74 ± 0.56
SW
64.87 ± 10.02
5.36 ± 1.28
1.69 ± 0.49
NM
52.73 ± 7.86
6.02 ± 1.43
1.51 ± 0.47
NW
48.34 ± 8.15
6.19 ± 1.55
1.47 ± 0.44
NR
44.65 ± 7.11
6.81 ± 1.68
1.42 ± 0.39
Table 2: Descriptive statistics (Mean ± SD) of behavioral parameters across five stimuli for the typical group.
Figure 2: Distribution of Mean Eye-Openness Percentage (EOP) across five stimuli for typical children. Human stimuli (SM, SW) show higher EOP, while robotic stimuli (NM, NW, NR) display reduced but sustained engagement.
Figure 3: Group-level behavioral response patterns across all stimuli.
Group
Stimulus
μ
σ
Full (%)
Partial (%)
None (%)
Typical
SM
1.59
0.66
68.38
22.22
9.40
Typical
SW
1.50
0.82
70.94
8.55
20.51
Typical
NM
1.41
0.77
58.12
24.79
17.09
Typical
NW
1.25
0.77
44.44
35.90
19.66
Typical
NR
1.28
0.82
51.28
25.64
23.08
Table 3: Descriptive statistics of mean response ( μ ), standard deviation ( σ ), and percentage distribution of response types across stimuli for typical children.
Group
Stimulus
μ
σ
Full (%)
Partial (%)
None (%)
Autistic
SM
0.78
0.88
29.17
19.44
51.39
Autistic
SW
0.96
0.85
33.33
29.17
37.50
Autistic
NM
1.56
0.67
65.28
25.00
9.72
Autistic
NW
1.56
0.73
69.44
16.67
13.89
Autistic
NR
1.74
0.56
79.17
15.28
5.56
Table 4: Descriptive statistics of mean response ( μ ), standard deviation ( σ ), and percentage distribution of response types across stimuli for autistic children.
Figure 4: Ridgeline density distributions showing within-group differences across stimuli.
Figure 5: Three-dimensional response surfaces illustrating mean response patterns across turns and stimulus conditions. (a) Typical group responses show stronger engagement with socially familiar cues. (b) ASD group responses demonstrate greater stability and higher engagement for robotic modalities across repeated turns.
Figure 6: Turn-wise response maps across the three name-calling trials. (a) Typical group responses. (b) ASD group responses.
Figure 7: Cronbach’s α values across stimulus conditions for the typical and autistic groups.
Dimension
Dominant in TD
Dominant in ASD
Interpretation
Response magnitude
Human voices (SM, SW)
Robot voices (NM, NW, NR)
Modality-specific sensitivity
Reliability ( α )
High for human
High for robot
Temporal consistency preference
Turn slope ( S )
Negative
Flat or positive
Habituation vs sustained engagement
Variance
Low for human
Low for robot
Selective stability
Signal discrimination ( d′ )
Low
High for robot stimuli
Diagnostic separability
Table 5: Summary of group-dominant metrics across analytical dimensions.
Stimulus
Metric
U
p -value
Cohen’s d
NM
Mean EOP
1210.5
<1×10−9
−1.08
NW
Mean EOP
1211.0
<1×10−9
−1.11
NR
Mean EOP
795.0
<1×10−14
−1.55
NM, NW, NR
Latency, Duration
ns
>0.16
–
Table 6: Between-group Mann–Whitney U analysis across stimuli.
Figure 8: (a) Latency and (b) Mean EOP distribution curves showing deviation from normality, justifying rank-based analyses.
Group
Metric
H
p -value
Key Contrast
Autistic
Duration
19.56
5.7×10−5
NR > NW
Autistic
Mean EOP
8.74
0.0126
NR > NW
Typical
Mean EOP
6.80
0.0334
NM > NR
Both
Latency
–
>0.40
not significant
Table 7: Within-group Kruskal–Wallis and post-hoc comparisons.
Figure 9: Raincloud plots showing Duration distributions across stimuli for autistic and typical participants. Longer durations under NR are evident for autistic children.
Pair
Group
ρ
p -value
Latency–Duration
Autistic
−0.47
4.8×10−12
Latency–Duration
Typical
−0.45
3.8×10−15
Duration–EOP
Autistic
0.28
5.5×10−5
Duration–EOP
Typical
0.35
1.9×10−9
Latency–EOP
Both
ns
>0.6
Table 8: Spearman correlations among behavioral metrics.
Variable
Autistic (range)
Typical (range)
Duration
0.726–0.825
0.668–0.780
Mean EOP
0.669–0.836
0.395–0.717
Latency
0.224–0.806
0.146–0.361
Table 9: Reliability coefficients (Cronbach’s α ) for behavioral metrics across groups.
Group
Variable
Stimulus NM
Stimulus NW
Stimulus NR
Interpretation
Autistic
Duration
−0.14
−0.40
+0.02
Duration slightly decreases for NM/NW but stabilizes under NR.
Autistic
Latency
−0.01
+0.01
−0.01
Essentially flat; almost no learning trend.
Autistic
Mean EOP
−0.07
−0.90
+0.56
Engagement declines slightly under NM,NW, but increases under NR.
Dance imitation integrates motor planning, sensorimotor integration, and social cognition, offering a sensitive framework to characterize motor behavior in autism. In this work, we explore a computational analysis framework to identify potential biomarkers that allow the design and development of improved medical and human-machine systems. We analyzed 3D motion capture data from autistic and neurotypical adults performing dance imitation under solo and socially-framed duo conditions. Methodologically, using Dynamic Time Warping, we quantified movement consistency and propose the Social Context Sensitivity Index (SCSI) to measure modulation of variability by social framing. These features were then used on a classifier to discriminate subjects into autistic or neurotypical groups. Results show that neurotypical adults exhibited increased movement variability in socially-framed imitation, especially in upper and lower limbs, whereas autistic adults maintained consistent movement across contexts. Classification achieved 79.2% balanced accuracy in distinguishing groups. These findings suggest that social context sensitivity in motor imitation constitutes a robust biomarker of autism-related motor behavior, highlighting the importance of social modulation in motor assessments and informing the development of inclusive human-centric technologies.
Lara Pereira, Teresa Sousa, Miguel Castelo-Branco +1
Institute of Systems and Robotics, University of Coimbra, Portugal · Coimbra Institute for Biomedical Imaging and Translational Research (CIBIT) of the University of Coimbra, Portugal · Intelligent Systems Associate Laboratory (LASI), Portugal +1
Robot gaze is a major component of human-robot dialogue coordination. Most studies of gaze in human-robot dialogue focus on face-to-face social conversations, but little is known about gaze in demanding task-focused interactions. In this paper, we investigate how the gaze of a robot game partner affects human visual attention and if humans tend to direct confirmation-seeking gazes towards the robot. In our study, we let participants play a collaborative word association game with a NAO robot acting as an embodied, LLM-driven conversational partner. Our experiments are conducted under two conditions, which implement mutual and referential gazes of the robot respectively. We record participants' gaze using eye tracking glasses and analyze the interactions using gaze coordinates, speech segments, key events and areas of interests. We find that robot gaze orientation does not affect the time to first fixation on words the robot proposed. We also find that participants gaze more often at the robot when their dialogue line contains confirmation requests, compared to when it does not. Our results indicate (likely also due to the cognitively demanding nature of the game) that the verbal aspect of this task overshadows the effects of referential robot gaze. These findings offer valuable insights for designing and validating robot gaze and turn-taking behavior in collaborative tasks which require coordination and efficient communication.
Jens V. Rüppel, Tim Schreiter, Andrey Rudenko +1
Chair of Perception for Intelligent Systems, Munich Institute of Robotics and Machine Intelligence (MIRMI), Technical University of Munich (TUM), Germany · Robotics Insitute Germany (RIG), Germany · Centre for Applied Autonomous Sensor Systems (AASS), Örebro University, Sweden
Social robots must coordinate speech and movement while interpreting multimodal human input. We present a multimodal framework that integrates speech, vision, and touch to generate contextually appropriate robot speech and movement. We evaluated the framework in virtual reality with 18 participants across five social interaction scenarios, comparing its performance against a human-teleoperated baseline. Although participants rated the autonomous system above the neutral midpoint for both naturalness and social appropriateness, the teleoperated baseline received significantly higher ratings on both measures. Behavioral coding and qualitative feedback revealed that conversational timing, clarification of ambiguous requests, and spatial coordination were important factors contributing to this performance gap. These findings highlight key challenges in developing autonomous robots capable of natural, multimodal social interaction.
Richard Xu, Dakarai Crowder, Alexis E. Block +1
University of Illinois at Urbana-Champaign, Champaign, IL, USA · Case Western Reserve University, Cleveland, OH, USA