Table 3.
Log likelihood (Bayesian information criterion, BIC) for exponential and hyperbolic discount functions.
| Experiment | Monkey D | Monkey J | ||
|---|---|---|---|---|
| Exponential | Hyperbolic | Exponential | Hyperbolic | |
| I-A | −1866.8 (3750.4) | −1990.5 (3997.8) | −2098.9 (4214.7) | −2197.0 (4410.9) |
| I-B | −2030.6 (4078.0) | −2138.6 (4294.0) | −1803.8 (3624.3) | −1887.4 (3791.7) |
| II-A | −1731.5 (3479.8) | −1566.7 (3150.2) | −2447.4 (4911.6) | −2319.9 (4656.6) |
| II-B | −1907.2 (3831.2) | −1742.9 (3502.6) | −2097.9 (4212.6) | −1927.8 (3872.5) |
| III-A | −1680.0 (3384.8) | −1895.6 (3816.0) | −1660.8 (3346.3) | −1700.1 (3424.9) |
| III-A* | −509.2 (1033.0) | −588.03 (1190.6) | −522.8 (1060.0) | −520.1 (1054.7) |
| III-B | −1464.6 (2946.0) | −1396.4 (2809.6) | −1720.4 (3457.6) | −1681.3 (3379.4) |
The discount functions with the better fit to the data are indicated by the BIC values in bold typeface. *indicates the results obtained from the Experiment III-A after excluding the trials with cyan disks.