Figure 4.

Mice can discriminate salt from sweet in a single lick (∼100 ms). a, Cumulative distribution of reaction times for correct-stop trials for mice discriminating between NaCl (stop) and sucrose (no-stop). Stimuli were 100 and 50 mm NaCl and 300 and 100 mm sucrose. The average reaction time was ∼275 ms, but mice were capable of stopping their licking to NaCl in a single lick (arrow), with reaction times of ∼100 ms, indicating that neural activity in this time window is sufficient for accurate taste-quality identification. b, Comparison of average rates of single-lick correct-stop trials and no-stop trial errors per session (mean ± SEM). Mice performed significantly more single-lick correct-stop trials per session than no-stop trial errors, indicating that single-lick correct-stop trials reflect true decisions by the animal and not anticipatory guessing (p < 0.00001, t test). c, Reaction times plotted against trial number show a significant negative correlation. This indicates that, as animals progressed further into a session, their reaction times became faster. This may be due to mice gaining more control over their licking behavior as they become less thirsty throughout a session. Most importantly, note the appearance of the majority of single-lick reaction times (arrow) toward the end of sessions. Single-lick reaction times, therefore, may represent the true sensory processing time needed to discriminate between NaCl and sucrose when motivation and motor control of licking are controlled for. d, Comparison of reaction times in the first versus the second half of sessions, demonstrating a 46% significant decrease in reaction time during the second half of sessions (mean ± SEM, p < 0.00001, t test).