Table 2.
Quantitative benchmark: f : frequency of attentional shifts [shifts/s], L : Latency time interval [s], U : normalized cpu utilization [%], D : data rate of input [Kbit/s], δ: duration of the acquisition[s].
| Experiment 1 | Experiment 2 | ||||
|---|---|---|---|---|---|
| Bright (~55.6 LUX) | Dim (~2.7 LUX) | ~27.2 LUX | |||
| iNVT | EVA | iNVT | EVA | EVA | |
| hor. top | 60.85% | 33.53% | 100% | 0% | . |
| hor. bot | 0% | 33.19% | 0% | 0% | . |
| ver. top | 39.15% | 15.06% | 0% | 0% | . |
| ver. bot | 0% | 18.21% | 0% | 100% | . |
| f (shifts/s) | 1.89 | 158.2 | 18.08 | 3.72 | 1708.80 |
| L (s) | (5.60 ± 0.3)e−2 | (23.2 ± 3)e−4 | (5.56 ± 0.3)e−2 | (23.1 ± 3)e−4 | (3.72 ± 1)e−3 |
| U (%) | 6.79 | 0.2 | 8.4 | 1.3 | 0.2 |
| D (Kbit/s) | 530e3 | 226.72 ± 0.078 | 530e3 | 20.32 ± 1.2 | 2.1e3 |
| δt (s) | 100 | 100 | 100 | 100 | 2.68 |
First experiment: number of hits clustered on the horizontally (top and bottom) and vertically (top and bottom) oriented grating stimuli under bright and dim illumination. Second experiment: performance of the EVA in details.