Fig 1.
Performance of a visual discrimination task.
(A) Schematic illustration of visual discrimination task. The mouse is head-restrained, on a running disk. Visual objects, displayed on a monitor, travel along the horizon from left to right. (B) Schematic illustration of the relative positions of mouse and monitor. (C) Running speed as a function of distance run by the mouse in two example trials. The position of the monitor, object and reward window are drawn to scale on the distance axis. (D) Summary of performance during a single example session, plot over time from the start to the end of the session (from left to right). Vertical lines indicate trials, sorted into four categories by object (rewarded vs unrewarded) and behavioral response (detected and undetected objects). The mouse collected rewards only on trials in which it indicated detection of a rewarded object (uppermost category). (E) Mouse-to-mouse and session-to-session variability in stop probability for rewarded and unrewarded objects and in discriminability (d'). Results are for 6 mice (rows) and across 6 sessions (columns).
Fig 2.
Performance across a range of contrasts.
(A) Mean running speed trajectories, for a single session, for objects of contrasts from 0 to 100%, rewarded (upper row) and unrewarded (lower row) objects. Dashed vertical lines: limits of reward window. (B) Psychometric curves for rewarded (vertical) and unrewarded (horizontal) objects for the session illustrated in panel A. Line, fit to Weibull distribution; error bars, 95% confidence intervals. (C) Discriminability of rewarded and unrewarded objects as a function of contrast. Each point is the mean (± SEM) from 6 mice.
Fig 3.
Behavioral artifact of deep brain illumination.
(A) Schematic illustration of the implanted guide cannula, fiber and deep brain illumination. Drawn approximately to scale. (B) Schematic illustration of the fiber implant (red circle) viewed from the dorsal surface of the head, illustrating the position of the fiber relative to the eyes. (C) Mean running speed trajectory for a single session in which 50% of trials included 640 nm illumination. Rewarded and unrewarded objects were presented at 100% contrast with (red traces) and without (black traces) 640 nm illumination. (D) Summary of performance during the session illustrated in panel B. Trials with and without 640 nm illumination are illustrated with red and black vertical bars, respectively. (E) Stop probabilities for the session illustrated in panel C. (F) Psychometric curves for a single wild-type mouse across four sessions, with different deep brain illumination in each session: no deep brain illumination (left panel), 10 mW of 473 nm illumination (center left), 10 mW of 589 nm illumination (center right), 10 mW of 640 nm illumination (right). Each session included trials with (colored symbols and lines) and without (black, grey) illumination. Stop probabilities for rewarded and unrewarded trials are illustrated with darker and lighter colors, respectively. (G) Stop probabilities for zero-contrast objects (false alarm rates) for 1, 3 and 10 mW at 473 (blue), 589 (yellow) and 640 nm (red) illumination. Results for each intensity and wavelength were collected in a different session and compared to the stop probability without illumination in the same session (in black). Asterisks denote significant effects of illumination (p < 0.01). Numbers of mice: 6, 8 and 8 mice for 1, 3 and 10 mW of 473 nm illumination; 6, 5 and 6 mice for 1, 3 and 10 mW of 589 nm illumination; 4, 5 and 5 mice for 1, 3 and 10 mW of 640 nm illumination. (H) Reward rate under illuminated and control conditions, within the same session. Results from individual mice are illustrated in grey, mean ± SEM of 5 mice in black. Rewards summed across rewarded and unrewarded objects of all contrasts; numbers of trials of each contrast with and without illumination were approximately equal.
Fig 4.
Deep brain illumination evokes activity in the left retina.
(A) Example ERG recordings from left and right eyes of a mouse during 473, 589 and 640 nm illumination (with no illumination from the visual stimulus monitor). Duration of illumination (200 ms) is indicated in grey. Amplitudes are normalized to those evoked by illumination from an external light source (200 ms, 20 mW, 473 nm illumination via the fiber tip mounted in front of the eye). (B) Peak ERG voltage (normalized as in panel B) as a function of optogenetic illumination intensity at 473, 589 and 640 nm.
Fig 5.
Measurement of light exiting the left eye.
(A) Schematic illustrating the experimental arrangement during measurement of light emitted through the left eye. Light (red arrow) propagated from the implanted fiber (red circle) to a spectroradiometer, placed in front of the left eye. (B) Intensities measured by the spectroradiometer during 10 mW illumination through the implanted fiber at 473, 589 and 640 nm. 3 mice.
Fig 6.
Light adaptation of left and right retinae.
(A) Schematic illustrating the experimental arrangement of mouse head, visual stimulus monitor and LED, the latter placed to illuminate the left retina. (B) Example ERG recordings from left and right eyes of a mouse under different ambient illumination conditions: monitor and LED off (dark-adapted; top row); monitor on and LED off (middle row), and monitor and LED both on (lower row). Optogenetic stimulus was 200 ms, 10 mW, 640 nm illumination (grey). (C) Summary of the effects of LED illumination on the peak amplitude of the ERG voltage in different adaptation states. Left point (dark-adapted) with monitor and LED off. Remaining points were acquired with the monitor on and the LED providing differing illumination intensities. Points represent mean ± SEM (3 mice). Arrowhead marks 0.14 Wsr-1m-2.
Fig 7.
Illumination of the left eye eliminates the behavioral artifact.
(A) Example of psychometric curves generated from single sessions without (left) and with (right) left retina adaptation (LED at 0.14 Wsr-1m-2). (B) Mean (± SEM) results and psychometric curves for 5 mice during left eye adaptation. (C) Summary of the change in false alarm rates with 10 mW, 640 nm illumination with and without left eye adaptation. 5 mice. Asterisk indicates p < 0.05, paired t-test.