The number of outcome- and response-selective units that showed particular combinations of outcome and response selectivity
| Selective for both outcome and response | Interaction with response selectivity | Interaction with outcome selectivity | Outcome-selective only | Response-selective only | |
|---|---|---|---|---|---|
| Outcome-selective units | |||||
| Odor presentation | 9 (26%)* | 2 (6%) | 24 (69%) | ||
| Response | 13 (31%) | 1 (2%) | 28 (67%) | ||
| Reward waiting | 7 (19%) | 2 (6%) | 27 (75%) | ||
| Response-selective units | |||||
| Odor presentation | 9 (17%)* | 6 (11%) | 38 (72%) | ||
| Response | 13 (14%) | 9 (10%) | 68 (76%) | ||
| Reward waiting | 7 (9%) | 7 (9%) | 66 (83%) |
The table shows the number of outcome- and response-selective units that showed particular combinations of outcome and response selectivity, namely, significant joint (outcome and response) selectivity (p < 0.05, by Mann–Whitney test) or outcome × response interactions (p < 0.05, by two-way ANOVA). Each column shows the numbers of units in each condition indicated by the column heading. The percentage of units in each cell out of all the units that were selective in the corresponding behavioral phase is shown in the parentheses. The numbers in the left column (Selective for both outcome and response) for both outcome- and response-selective units are from identical units. Only during the odor presentation phase were there significantly more neurons that encoded both outcome and response than would be expected by chance. *p = 0.035 for the binomial test for mutual independence of outcome and response selectivity in each phase (for details, see Materials and Methods).