Article Figures & Data
Figures
- Figure 1.
Timeline of events in the touchscreen-based orienting task during baseline training with a difficulty level of 5 s of stimulus presentation. The house light is turned off by default and a sugar pellet (reward) is provided to prime the session, along with illumination of the food magazine. A trial begins as soon as the animal enters the food magazine (0 s), after which the food magazine light is switched off. Then, the stimulus is presented after 5 s, and the animal has to make a response (nose-poke) within a 5-s limited hold of stimulus presentation. A, In a correct trial, the animal has correctly poked the illuminated panel and is rewarded immediately. The house light remains off throughout the trial. B, In an incorrect trial, the animal has poked a non-illuminated panel and receives no reward. The house light turns on immediately after the incorrect response for a period of 5 s to signal a timeout and discourage the animal for an inappropriate behavior. C, In an omission trial, the animal has not poked any panel at all and thus is not rewarded. House light turns on for 5 s right after the limited hold of stimulus presentation (5 s) to signal a timeout. For all trials: the next trial begins after a 5-s ITI with illumination of the food magazine. Note that whenever a sugar pellet (reward) is dispensed, it is accompanied by illumination of the food magazine light, which is then switched off once the animal collects the reward.
- Figure 2.
Description of the three-choice orienting protocol. A, After completing all the pretraining steps, the rats undergo baseline attention training, where they learn to expect the center stimulus trial as the most likely trial type, by having the center stimulus appear in 80% of the trials while the left and right stimuli each appear in only 10% of the trials. Animals are rewarded with one sugar pellet on correctly nose-poking the illuminated panel and punished with a timeout and inversion of the house light for poking a non-illuminated panel. The duration of the stimulus presentation is decreased from 30 to 1.5 s through seven stages of baseline training to increase task difficulty. B, Upon passing the criteria, rats perform the testing session after injections of CNO or vehicle. Testing sessions include baseline trials to reinforce the basic strategy the rats must use, in addition to ambiguous trials where flanking stimuli are presented with varying SOAs to test their orienting bias at various difficulties; 75% of the test trials are center-only trials, where only the center stimulus is presented for 5 s. The remaining 25% are double-stimulus trials where the left and right stimuli appear either simultaneously or with a delay (0. 5 or 1 s) with left stimulus preceding the right or vice versa.
- Figure 3.
DREADD expression in the SC across all DREADD-expressing animals. Expression is displayed across six slices, from bregma −5.52 mm as the most rostral, to bregma −6.60 mm as the most caudal slice. The color gradient on the SC represents the number of animals that had DREADD expression at each subregion of the SC: the darker the gradient at a subregion, the higher the number of animals that had DREADD expression in that subregion. Each schematic slice is accompanied by an immunohistochemistry photograph of the same slice from a representative animal. The images of the slices were taken from The Rat Brain Atlas in Stereotaxic Coordinates from Paxinos and Watson (2006).
- Figure 4.
Body turning but not head turning behavior is increased following CNO in DREADD but not sham animals. Rats spontaneously explored an enclosed arena 20 min after injection of vehicle or CNO. A difference score was calculated for each measure of interest between CNO and vehicle for each animal, and the sham and DREADD groups were compared using independent t tests or Mann–Whitney U test (#). A, There was no difference in the CNO-vehicle score of total distance traveled. B, No difference in total 360° rotations. C, No difference in percentage clockwise 360° rotations. D, No difference in the sum of all head turn angles made during the 20-min test (clockwise head turns were denoted as positive and anticlockwise head turns as negative). E, There was a significant increase in the CNO-vehicle score of the turn angle sum (clockwise body turns were denoted as positive and anticlockwise body turns as negative). All values shown are mean ± SEM, with individual dots representing individual animals; * significant effect with p < 0.05, $ denotes adjusted p value based on lack of homogeneity of variance. For more information on how turn angles were measured, see Extended Data Figure 4-1.
- Figure 5.
Performance and motivation are stable across pretest days, but animals omit fewer trials and respond more quickly by the last pretest day. Animals underwent baseline training and were then tested up to four times. These graphs show performance parameters on the days before testing days for both sham and DREADD animals combined, which was analyzed using repeated measures ANOVA or nonparametric Friedman test (#). A, Overall accuracy did not change across pretest days. B, Percentage omission significantly decreased across pretest days. C, Reward collection latency was unchanged across pretest days, indicating stable performance and motivation across pretest days. D, Correct response latency significantly decreased across pretest days, adjusted p refers to comparison between day 1 and day 4, which was adjusted for multiple comparisons. These results indicate that by undergoing test sessions which include more reward and less punishment, animals learned to omit fewer trials and respond more quickly. All values shown are mean ± SEM, with light gray lines representing individual animals; * significant effect with p < 0.05. For more information on progression through all training stages, see Extended Data Figure 5-1.
- Figure 6.
Animals show strong center bias on pretest days. Data from all pretest days were combined to analyze trial-type-specific performance using repeated measures ANOVA or nonparametric Friedman test (#). All p values shown are adjusted for multiple comparisons. A, Animals were more accurate in response to center trials. B, The omission rate was higher to left side trials than to center trials. C, Correct responses to center trials were faster than to side trials. All values shown are mean ± SEM, with individual dots representing individual animals; * significant effect with p < 0.05.
- Figure 7.
CNO reduces center bias in DREADD but not sham animals during testing. Test sessions included 75% center trials, similar to the ones shown during baseline training. A CNO-vehicle score was calculated for each parameter of interest. Depending on the distribution, an independent t test or Mann–Whitney U tests (#) were conducted to compare sham and DREADD animals. A, Left Responses were unchanged. B, There was a significant decrease in the percentage of correct center responses and (C) a trend to an increase in rightward responses. D, There was also a significant increase in the percentage of omissions in DREADD animals. E, A reduction in center bias can also be seen by an increase in the time it took DREADD animals to respond correctly to the center following CNO. All values shown are mean ± SEM, with individual dots representing individual animals; * significant effect with p < 0.05.
- Figure 8.
CNO produces a strong rightward bias in DREADD animals. A, B, During test sessions, SOA trials were used to assess the extent of rightward bias produced by CNO in DREADD animals. Circles in A, B indicate the average proportion of right responses at that particular SOA for the entire group with either vehicle (blue) or CNO (red). A logistic function was fitted to the proportion of rightward responses for each animal, as well as for the whole group. DREADD animals, but not sham animals, showed a left shift of their response profile after CNO injection. C, The PES (0.5 right responses) was extrapolated from each individual animal’s curve following either vehicle or CNO injection. A value of −1 was used as a conservative estimate of PES for animals where the logistic function revealed an extreme rightward bias that could not be extrapolated to a PES. A two-way ANOVA was conducted to compare the PES within each group, which showed a significant decrease in PES in the DREADD group, but no change in the sham group. Values are mean ± SEM, with individual dots representing individual animals; * significant effect with p < 0.05. For more information of each single animal’s performance, see Extended Data Figure 8-1 for sham animals and Extended Data Figure 8-2 for DREADD animals.
Movies
- Movie 1.
ANYmaze video tracking with adjusted settings consistently track the head and body position. Video sample from the OFT, showing ANYmaze tracking using the adjusted settings described in the methods section. The video sample shows consistent tracking of the animal’s head (green) and center (orange) positions, which are used to calculate turning and rotation measures as described in the methods section and supplemental figures.
- Movie 2.
Baseline center training leads to a strong center bias. Video sample from the touch screen orienting task baseline training. The video sample shows the animal orienting towards the center before the stimulus is shown and correctly responding to the center stimulus and receiving a reward (00:00–00:02 for orienting before a stimulus is shown, 00:02–00:04 for responding to the stimulus). A similar sequence of events can also be seen in the following three trials (00:12–00:15, 00:28–00:31, 00:40–00:44). Following a correct response, the animal collects its reward and initiates a new trial. The last trial in the video shows an incorrect side trial, where the stimulus was shown on the side, but the animal’s center bias made it incorrectly choose the center (00:51–00:54).
Extended Data 1
ABET II Touch software (Lafayette Instruments). Download Extended Data 1, ZIP file.
Extended Data Figure 4-1
Turn measure examples. Visual depiction of turn angle measures recorded by ANYmaze software. A, Turn angle: a vector of movement from one position of the animal’s center point to the next is created. For each vector, the angle between it and the previous vector is calculated with anti-clockwise movement being negative and clockwise movement being positive (ANYmaze). B, Head turn angle: for each position of the animal’s head, a vector is created from the animal’s center point to the head. The angle between this vector and the same vector for the previous position of the animal’s head is calculated, with anti-clockwise movement being negative and clockwise movement being positive. C, Example calculation from successive head turns: panels are numbered from left to right as C1, C2, C3, C4. C1 to C2 head turn angle = +75; C2 to C3 head turn angle = +45; C3 to C4 head turn angle = –35. Turn angle sum is the sum of all turn angles, measured as in panels A, B, in a particular time period (e.g., the head turn angle sum for time period C1–C4 = 75 + 45 – 35 = +85). To complete a full rotation, an animal must make consecutive turns that add up (cumulative turn angle) to 360°. Reversal (turning towards the opposite direction) at any point before the cumulative angle reaches 360° resets the cumulative angle to 0 and starts adding up successive turns to the opposite direction (e.g., considering panels C1–C4, the cumulative angle counted towards a full rotation is –35 because the animal made a reversal, despite the head turn angle sum being +85). Because of the effects of the reversal, we believe the turn angle sum is a more accurate representation of turning behavior than complete 360° rotations. Download Figure 4-1, EPS file.
Extended Data Figure 5-1
Time in training stages. Rats progress quickly through all training stages in the five-choice orienting task. Pretraining stages (habituation to must initiate) acclimate rats to the chambers and them in a stepwise manner to associate nose-poking stimuli on the screen with reward. All rats passed quickly through pretraining stages, taking at most 4 d to reach criterion and an average of about 5 d to finish all of pretraining. Baseline training stages involve training the rats to orient towards the center panel, with a stimulus being shown 10% in the left panel, 80% in the center panel, and 10% in the right panel. To pass, rats must achieve an overall accuracy of 80% with fewer than 20% omitted trials. Rats progressed quickly through baseline 1 (60-s stimulus duration) to baseline 4 (10-s stimulus duration). Even on the most difficult stage, rats took an average of 3 d to reach criterion and just under 12 d to complete all baseline stages. Download Figure 5-1, EPS file.
Extended Data Figure 8-1
No rightward bias in individual sham animals. Sham animals do not show a rightward bias in double stimulus trials following CNO injection. Individual psychometric curves for sham animals showing their performance on double stimulus trials within the SOA range –1 to +1 following either vehicle (blue) or CNO (red) injections. A rightward bias would manifest as a leftward shift in the psychometric curve, indicating a higher proportion of rightward responses. As can be seen by the individual curves, sham animals did not exhibit a rightward bias following CNO injection compared to vehicle. Download Figure 8-1, EPS file.
Extended Data Figure 8-2
Rightwards bias in individual DREADD animals. DREADD animals show a rightward bias in double stimulus trials following CNO injection. Individual psychometric curves for DREADD animals showing their performance on double stimulus trials within the SOA range –1 to +1 following either vehicle (blue) or CNO (red) injections. Animals are numbered left to right, top to bottom (row 1: animals 1–4; row 2: animals 5–8, row 3: animal 9). A rightward bias would manifest as a leftward shift in the psychometric curve, indicating a higher proportion of rightward responses. Several rats showed extremely strong rightward bias, as indicated by almost exclusive rightward responses regardless of SOA (animals 3, 5, 8, and 9). Other animals showed a more modest rightward bias (animals 2, 4, and 6) and only two rats showed no rightward bias following CNO (animals 1 and 7) Download Figure 8-2, EPS file.
In this issue
