Carl Westin

4-3 Method 71 TWIN AKON MEMS HOOKS LANG DAMS GYRO RIVET Source: Human Advisory agreement Neutral Agree Disagree Accept Reject 20 Time: Score: Run: 00:01:30 84.98% 1 / 5 QS1338 270 260 RIVET QS1338 270 260 RIVET PA5424 270 310 MEMS OM2514 270 200 RIVET WY8047 270 200 AKON MU2653 270 310 GYRO OM3185 270 260 GYRO DT4511 270 310 DAMS UG7514 270 320 GYRO ZS8580 270 310 TWIN NM3867 270 310 TWIN SM7071 270 310 RIVET NG8781 270 320 TWIN RA4743 270 300 RIVET SG3047 270 300 RIVET NA9895 270 310 AKON ZM9694 270 310 AKON PG4310 270 320 AKON AE1870 270 320 AKON OS2071 270 320 AKON YY7466 270 300 TWIN MA1538 270 310 TWIN F IGURE 4-2: The simulator interface showing a resolution advisory in the heading band SSD for QS1338. The advisory suggests a right turn and increase in speed (320 knots). In this example, the adviser is human, as indicated by the advisory dialog window in the upper right corner. The same window enables the controller to either accept or reject the advisory, but first after providing an agreement rating. Time to advisory expiration is shown in the bottom of the window (20 seconds). The box in the upper left corner shows time remaining of scenario, current performance score, and run out of total runs. Aircraft depicted in lighter grey are off track and require a vector to their assigned exit point (indicated in the bottom field of the aircraft label). 4-3-5 Independent variables A within-participant study was conducted with advisory conformance ( cfY = con- formal and cfN = nonconformal) and advisory source ( Hum = human and Aut = automation) as independent variables. Conformal advisories were individually- tailored solutions based on a participant’s conflict solving style. Nonconformal ad- visories consisted of another participant’s style that was different but solved the con- flict. Participants were not informed of the conformance manipulation. Advisory conformance was varied within participants and within source conditions according to a Latin Square matrix. Advisory source was manipulated by framing. Participants were provided with different source descriptions in briefings prior to each conditioned simulation run.