Table 1.

Overview of the interactions used to study human-human interaction to date and the resulting outcomes

Principal findings	Papers
Experimental paradigms
1. HHI research has predominantly focused on sensorimotor collaboration, not cooperation
Sensorimotor collaborations	Reed [56, 65], Groten [70], Melendez-Calderon [43], Ikeura [50], Rahman [51], Basdogan [52], Sallnas [53], Gentry [54, 64], van der Wel [55], Feth [66]
Sensorimotor cooperation	Ikeura [60], Galvez [86]
2. The majority of HHI research has used visuomotor tasks with limited degrees-of-freedom
Constrained visuomotor tasks	Melendez-Calderon [43], Ikeura [50], Rahman [51], Basdogan [52], Sallnas [53], Gentry [54, 64], van der Wel [55], Reed [56]
Whole-body non-visuomotor tasks	Galvez [86]
3. In most HHI research, specific roles for each member of a dyad are rarely defined ahead of time
Unassigned roles	Reed [65], Groten [70], Melendez-Calderon [43], Rahman [51], Basdogan [52], Sallnas [53], Gentry [54, 64], van der Wel [55], Reed [56, 65], Feth [66]
Assigned roles	Ikeura [50, 60]
Experimental outcomes
4. Dyads typically perform as well or better than either member of a dyad alone
Superior dyad performance	Reed [65], Gentry [54], Feth [66]
Equivalent individual and dyad performance	van der Wel [55]
5. The addition of haptic feedback improves dyad performance compared to visual feedback alone
	Basdogan [52], Sallnas [53], Gentry [64], Groten [67, 71]
6. Members of a dyad apply higher forces than during either of their individual performances
	van der Wel [55], Reed [68], Groten [70, 71], Feth [66]
7. Members of a dyad spontaneously assume specific roles, performing portions of a joint motor task
	Melendez-Calderon [43], Reed [32, 56]
Starting or ending movement	Reed [32, 68]
Adding or absorbing energy	Feth [66]