PSD-95 KO mice exhibit deficits in prey capture behaviour

(A) Schematic of the arena used for testing prey capture behaviour. Behaviour was recorded at 60 fps using an overhead camera (for details see Methods). (B) Experimental timeline with relevant phases indicated by grey lines; after 3 days of handling/acclimatisation (A1-A3), mice were food deprived and made to hunt crickets in the arena until they exhibited stable capture performance (D1-Dn-1). The following two test days consisted of mice hunting crickets first with intact binocular vision (Dn), after which one eye was sutured shut to allow monocular testing in the same animals ∼20-24 hrs later (Dn+1). (C) Schematic representation of predatory behaviour and parameters recorded. The first schematic depicts the time to 1st approach, i.e. the time it takes for the mouse to detect the cricket and make its first approach. The second schematic depicts the azimuth between mouse and cricket, i.e. the head position of the mouse in relation to the cricket. The third schematic depicts an approach, defined as an epoch where the mouse speed was >6 cm/s, mouse-cricket distance decreased by a rate of 7 cm/s and the azimuth was <150 degrees. The fourth schematic depicts a mouse-cricket contact, defined as a mouse-cricket distance <3 cm and azimuth <90°. (D) Probability to catch crickets on a given day within 10 minutes of prey exposure. Individual data points are overlaid on the mean±SEM lines for each genotype. (E) Prey capture times of PSD-95 WT and KO mice on the last training day (Dn-1). Each point represents the median capture time of individual mice whereas the bars indicate the mean±SEM. (WT: 11.8±1.8 s, n=8; KO: 57.2±10.9 s, n=7; Mann-Whitney, p= 0.0003). (F and G) Comparison of median prey capture times for individual PSD-95 WT and KO mice between the last training day and the binocular test day, respectively (WT: Dn – Dn-1 = -3.6±1.71 s, n=8; Wilcoxon’s test, p= 0.055; KO: Dn – Dn-1 = -5.6±2.7 s, n=7; Wilcoxon’s test, p= 0.125).

Monocular PSD-95 KO mice hunt faster than with intact binocular vision while WT mice get worse with monocular vision

(A) Example tracks of the same PSD-95 WT and KO mice under binocular and monocular condition, respectively, until successful capture (scale bar, 10 cm). (B) Percentage change in median capture times of individual WT/KO mice between binocular and monocular condition (WT: 169.2±70.5 %, n=8; KO: -12.7±24.8 %, n=7; Mann-Whitney test, p=0.006). (C) Within-group comparisons of prey capture times of WT and KO mice under both binocular and monocular (hatched bars) conditions, respectively. Each datapoint represents a trial and each bar represents the mean±SEM (WT binocular: 8.9±0.9 s, n=31; WT monocular: 22.2±3.6 s, n=32; KO binocular: 74.2±13.9 s, n=27; KO monocular: 42±5.6 s, n=34). ○○, binocular ○●, monocular. Linear mixed effects model with estimated marginal means for multiple comparisons (see Table 1 and S1). **p< 0.01 and ****p < 0.0001.

Statistics of prey capture parameters in the light trials

Summary of the statistical outcomes of the type II Wald χ2 tests for each prey capture parameter. A complete overview of the statistical results can be found in Supplementary Table S1. Abbreviations: n.s. = non-significant; tr. = travelled.

Binocular PSD-95 KO mice exhibit deficits in diverse epochs of prey capture behaviour but partially improve with monocular vision

(A) Example traces of mouse-cricket distance and speed over time until successful capture for an individual WT (black lines, left two panels) and KO (blue lines, right two panels) mouse with binocular and monocular vision, respectively. Approach epochs are labelled in red. (B) Rate of approaches per second towards prey. WT binocular: 0.27±0.02, n=25; WT monocular: 0.2±0.02, n=31; KO binocular: 0.16±0.02, n=27; KO monocular: 0.15±0.01, n=34. (C) Latency to initiate the first approach sequence towards prey. WT binocular: 1.38±0.35 s, n=25; WT monocular: 3.89±0.75 s, n=31; KO binocular: 8.37±3.13 s, n=27; KO monocular: 2.84±0.65 s, n=34. (D) Probability of a contact given a successful approach. WT binocular: 0.85±0.05, n=25; WT monocular: 0.76±0.04, n=31; KO binocular: 0.58±0.05, n=27; KO monocular: 0.6±0.05, n=34. (E) Probability of capturing prey given a successful contact. WT binocular: 0.51±0.06, n=30; WT monocular: 0.38±0.05, n=32; KO binocular: 0.17±0.02, n=27; KO monocular: 0.21±0.02, n=34. (F) Duration of each contact with prey. WT binocular: 1±0.12 s, n=31; WT monocular: 0.85±0.12 s, n=32; KO binocular: 2.17±0.5 s, n=27; KO monocular: 1.42±0.16 s, n=34. (G) Total duration of contact with prey. WT binocular: 2.35±0.24 s, n=31; WT monocular: 2.88±0.45 s, n=32; KO binocular: 14.09±1.87 s, n=27; KO monocular: 7.24±0.72 s, n=34. Bars represent mean±SEM. Each data point corresponds to a trial. Data points in E and F correspond to the mean value in a single trial. Bars of WT and KO mice are indicated by black and blue, respectively. ○○, binocular ○●, monocular. Linear mixed effects model with estimated marginal means for multiple comparisons (see Table 1 and S1). *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001.

PSD-95 KO mice exhibit inefficient appetitive locomotion but improve with monocular vision

(A) Percentage of time mice stay immobile at baseline (non-approach phases). WT binocular: 2.35±0.24%, n=31; WT monocular: 2.88±0.46%, n=32; KO binocular: 14.09±1.87%, n=27; KO monocular: 7.24±0.72%, n=34. (B) Percentage of time mice are in an arrest-like state during an active approach towards prey. WT binocular: 2.58±0.61%, n=25; WT monocular: 1.99±0.38%, n=31; KO binocular: 2.48±0.44%, n=27; KO monocular: 3.04±0.5%, n=34. (C) Distance travelled at baseline when mice are not actively approaching prey. WT binocular: 92.96±12.99 cm, n=25; WT monocular: 167.3±16.19 cm, n=31; KO binocular: 388.5±40.25 cm, n=27; KO monocular: 260.6±24.8 cm, n=34. (D) Distance travelled during approach epochs towards prey. WT binocular: 71.74±8.92 cm, n=25; WT monocular: 110.5±11.68 cm, n=31; KO binocular: 139.4±12.39 cm, n=27; KO monocular: 133.2±10.68 cm, n=34. (E) Speed of mice at baseline (non-approach phases). WT binocular: 10.54±1.36 cm/s, n=25; WT monocular: 10.65±1.24 cm/s, n=31; KO binocular: 6.81±0.8 cm/s, n=27; KO monocular: 7.97±0.59 cm/s, n=34. (F) Speed of mice during the active approach phases. WT binocular: 19.86±1.22 cm/s, n=25; WT monocular: 21.55±1.1 cm/s, n=31; KO binocular: 15.01±0.73 cm/s, n=27; KO monocular: 14.64±0.63 cm/s, n=34. (G) Speed of mice around ±1 s of contact with prey. Lines and shaded areas indicate mean±SEM. Dashed line indicates the time of a contact. Bars indicate mean±SEM. Each data point in A and B corresponds to % of immobility in each trial. Data points in E and F correspond to the median mouse speed in a trial. ○○, binocular ○●, monocular. Linear mixed effects model with estimated marginal means for multiple comparisons (see Table S1). *p<0.05, ***p<0.001.

Binocular visual field bias is conserved in KO mice irrespective of visual condition (binocular/monocular), unlike in WT mice

(A) Mean absolute azimuth as a function of the mouse-cricket distance until contact during approach phases. (B) Polar probability histograms (10° bins) of azimuth at the end of approach towards prey. Dashed red line indicates the binocular visual field (±20°). (C) Absolute azimuth at the end of approach towards prey. WT binocular: 14.4±2.62°, n=49; WT monocular: 26.8±2.42°, n=97; KO binocular: 14.5±1.61°, n=140; KO monocular: 17.6±1.9°, n=148. Bars indicate mean±SEM. ○○, binocular ○●, monocular. Linear mixed effects model with estimated marginal means for multiple comparisons (see Table 1 and S1). *p<0.05, **p<0.01.

PSD-95 KO mice exhibit poorer orientation discrimination compared to WT mice but improve monocularly

(A) Schematic of the visual water task adapted for testing orientation discrimination threshold (panel A is reproduced from Figure 8 from Favaro et al., 2018). (B) Total number of trials required for training WT and KO mice before beginning the test phase. WT: 103±24, n=10; KO: 153±24, n=9; Mann-Whitney, p=0.016. (C) Orientation discrimination thresholds of WTs and KOs with binocular and monocular vision during the test phase. WT binocular: 17±2°, n=8; WT monocular: 34±6°, n=6; KO binocular: 52±4°, n=11; KO monocular: 32±2°, n=4. Multiple linear regression with estimated marginal means for multiple comparisons. F3,25=14.17, p<0.0001; βg= -35.45, p<0.0001; βv=-20.52, p=0.0065; βg*v=37.53, p=0.0005. WTb - WTm: t(25)=2.66, p=0.018; KOb - KOm: t(25)=2.97, p=0.013; WTb – KOb: t(25)=6.45, p<0.0001; WTm - KOm: t(25)=0.273, p=0.787. Bars indicate mean±SEM. ○○, binocular ○●, monocular. *p<0.05, ****p<0.0001.