Visual Dominance and Visual Egalitarianism: Individual and Group-Level Influences of Sex and Status in Group Interactions

Abstract

This study investigated visual dominance and visual egalitarianism of men and women (N = 94; 17 teams) in team meetings at diverse workplaces. Two novel gaze-related measures were developed: (a) a group visual dominance ratio (group-VDR) assessing each member’s visual dominance vis-à-vis all other members, and (b) a gaze distribution index (GDI) assessing each member’s visual egalitarianism to all group members. Multilevel analyses were conducted to account for influences of the team members’ sex and status on the individual level and for influences of sex and status composition of the teams, and the team leaders’ sex on the group level. Results suggested that high-status individuals displayed more visual dominance than low-status individuals. The significant interaction of individuals’ sex and status indicated that the positive relationship of status and visual dominance applied particularly to women. The more women in a team, the more visual dominance was displayed. The team leader’s sex significantly influenced visual egalitarianism: Gaze distribution was less egalitarian when the team leader was male.

Appendices

Appendix A

Details on the Development of the Group-VDR

Example: For a group of three persons, Formula 2 would be applied as follows (lwt = looks while talking; lwl = looks while listening; gr = group):

$$ {\frac{{{\frac{{{\text{A}}\,{\text{lwt}}\,{\text{B}} + {\text{A}}\,{\text{lwt}}\,{\text{C}} + {\text{A}}\,{\text{lwt}}\,{\text{gr}}}}{{{\text{A}}\,{\text{talk}}\,{\text{B}} + {\text{A}}\,{\text{talk}}\,{\text{C}} + {\text{A}}\,{\text{talk}}\,{\text{gr}} + {\text{A}}\,{\text{talk}}\,{\text{away}}}}} \times 100}}{{{\frac{{{\text{A}}\,{\text{lwl}}\,{\text{B}} + {\text{A}}\,{\text{lwl}}\,{\text{C}} + {\text{A}}\,{\text{lwl}}\,{\text{gr}}}}{{{\text{B}}\,{\text{talk}}\,{\text{A}} + {\text{B}}\,{\text{talk}}\,{\text{C}} + {\text{B}}\,{\text{talk}}\,{\text{gr}} + {\text{B}}\,{\text{talk}}\,{\text{away}} + {\text{C}}\,{\text{talk}}\,{\text{A}} + {\text{C}}\,{\text{talk}} + {\text{B}} + {\text{C}}\,{\text{talk}}\,{\text{gr}} + {\text{C}}\,{\text{talk}}\,{\text{away}}}}} \times 100}}} $$

Missing data due to occasional lack of visibility of some group members and their absence from part of the meeting (e.g., for an incoming phone call) caused us to adjust the group-VDR calculations in two steps: (a) to adjust for lack of visibility, we used the following formula (Formula 4; group-VDR considering out-of-sight times):

$$ {\frac{{{\frac{{{{A}}\,{{lwt}}\,{{total}} - ({{A}}\,{{lwt}}\,{{away}} + {{away}}\,\% \,{{of}}\,{{lwt}}\,{{without}}\,{{sight}})}}{{{{A}}\,{{talk}}\,{{total}}}}} \times 100}}{{{\frac{{{{A}}\,{{lwl}}\,{{total}} - ({{A}}\,{{lwl}}\,{{away}} + {{away}}\,\% \,{{of}}\,{{lwl}}\,{{without}}\,{{sight}})}}{{{{talk}}\,{{total}}\,{{group}}\,{{without}}\,{{A}}}}} \times 100}}}, $$

(b) when out-of-sight times and absence times occurred simultaneously, the following formula provided an adequate solution (Formula 5; group-VDR considering out-of-sight and absence times)

$$ {\frac{{{\frac{{{{A}}\,{{lwt}}\,{{total}} - ({{A}}\,{{lwt}}\,{{away}} + {{away}}\,\% \,{{of}}\,{{lwt}}\,{{without}}\,{{sight}})}}{{{{A}}\,{{talk}}\,{{total}}}}} \times 100}}{{{\frac{{{{A}}\,{{lwl}}\,{{total}} - ({{A}}\,{{lwl}}\,{{away}} + {{away}}\,\% \,{{of}}\,{{lwl}}\,{{without}}\,{{sight}})}}{{{{talk}}\,{{total}}\,{{group}}\,{{without}}\,{{A}} - {{\text{talk}}}\,\% \,{{\text{total}}}\,{{\text{group}}}\,{{\text{in}}}\,{{\text{absence}}}\,{{\text{time}}}({{\text{s}}})}}} \times 100}}} $$

Ad Formula 4: Some persons were sometimes covered by other persons. In these cases we coded the speaking mode (lwt or lwl), yet, the gaze direction, i.e., the target person looked at, needed to be inferred. Presupposing that persons show a similar gaze behavior when they are not visible compared to when they are visible, we chose Formula 4. For example, to calculate ‘away % of lwl without sight’ for person C from team X we proceeded in the following way: person C from team X was in lwl-mode for 14,604 frames. She looked away for 2,715 frames. For 465 frames she was out of sight (oS). The total time that C was in lwl-mode and at the same time visible, thus, amounted to 14,604 − 465 = 11,889 frames. In about 23% of this time she looked away (away = 2,715). Assuming that she would display a similar gaze pattern for the time that she was out-of-sight, we presupposed that she would look away for 23% of the out-of-sight-time as well, which amounted to 106 frames. This value was added to the total observed lwl-away-time and then subtracted from the total listening-time. Lwt-values were treated respectively.

Ad Formula 5: Further modifications in calculating the group-VDR were necessary when person A joined the team meeting at a later point, left at an earlier point, or intermittently left the room, for example, for a phone call. These absence times needed to be taken into account for group-VDR calculations for the absent person in the value ‘talk total group without A’. The assumption was that the lwt-behavior of the whole group during the absence times is proportional to the lwt-behavior of the whole group during the entire observation time. The computations changed as shown in Formula 5.

We calculated the total value of group’s lwt without A in proportion to the entire time of a session in Formula 4. This percent value was the value multiplied with A’s out-of-sight time. The result was the estimator for A’s lwt time while out of sight which was subtracted from the overall values. Hence, the resulting value ‘talk total group without A’ considered the out-of-sight times. When out-of-sight and absence times occurred both at a time Formula 5 provided an adequate solution. Depending on the circumstances encountered, the appropriate formula needs to be chosen (with formula parts in italics flexibly applied).

Technical specifications: A frontal view of all participants allows to compute the group-VDR as precisely as possible. One overall perspective is preferable to split-screen taping, because it increases target accuracy.

Appendix B

Mathematical Derivation of the Gaze Distribution Index (GDI) for a Person p

We calculated the GDI-value of a person as the sum of the absolute differences of this person looking at each other team member and the maximum balanced gaze distribution (i.e., each team member is looked at for the exact same amount of time). The most unbalanced value would result for a person looking at only one person during the entire time. This value, however, is still influenced by group size. In order to get a standardized GDI _p -value we calculated the maximum polarized gaze distribution for each possible group size using the Manhattan norm, and divided the empirically derived values by this maximum value for standardization purposes. Resulting GDI _p -values were reversed in polarization and lay then between 0 polarized and 1 balanced.

Note: In the following mathematical derivation n = number of persons in the team; t _i is the standardized time that the person p looks at another person p _i , for i = 1,…, n − 1. (n − 1 being the standardized time)

Each person p is assigned a GDI-value GDI _p (i.e., a standardized measure for the gaze distribution of the person p) with the following attributes:

$$ 0 \le {\text{GDI}}_{p} \le 1 $$

• a. GDI _p  = 1 for balanced gaze distribution, i.e., each other person in the team is looked at for the exact same amount of time: ${\frac{1}{n - 1}}$

• b. GDI _p  = 0 for maximum polarized gaze distribution, i.e., person p looks at only one person during the entire time;

To obtain a standardized measure of egalitarianism we used the Manhattan norm

$$ \sum\limits_{i = 1}^{n - 1} {\left| {{\frac{1}{n - 1}} - t_{i} } \right|} $$

(the sum of the absolute differences of ${\frac{1}{n - 1}}$ as the balanced gaze distribution and t _i the actual standardized gazing time). For maximum balanced gaze results

$$ \sum\limits_{i = 1}^{n - 1} {\left| {{\frac{1}{n - 1}} - {\frac{1}{n - 1}}} \right| = 0,} $$

and for maximum polarized gaze, i.e., only one person is gazed at for the entire time:

$$ \left| {{\frac{1}{n - 1}} - 1} \right| + \sum\limits_{i = 1}^{n - 2} {\left| {{\frac{1}{n - 1}} - 0} \right| = 1 - } {\frac{1}{n - 1}} + {\frac{n - 2}{n - 1}} = 2 \times {\frac{n - 2}{n - 1}}. $$

A standardized GDI-value (between 0 and 1), fulfilling the conditions (a) to (c), results from

$$ GDI_{p} = 1 - {\frac{{\sum\limits_{i = 1}^{n - 1} {\left| {{\frac{1}{n - 1}} - t_{i} } \right|} }}{{2 \times {\frac{n - 2}{n - 1}}}}} = 1 - {\frac{n - 1}{2(n - 2)}} \times \sum\limits_{i = 1}^{n - 1} {\left| {{\frac{1}{n - 1}} - t_{i} } \right|} $$

After reversal of polarities, GDI-values lie between 0 polarized and 1 balanced.