Fig. 2

UK4 enhances fear-dependent contextual memory yet decreases long-term spatial learning. Fear-dependent memory and learning were assessed using fear conditioning. At phase 1—acquisition, mice were presented with a tone, followed by a foot shock. Cue-associative learning was assessed by measuring freezing behaviour during the presentation of the tone (a), whereas context-associative learning was determined in-between tones (b). At phase 2—cued memory, mice received 40 presentations of the same cue (the first 10 are shown), without foot shock, in a different context, in which fear-dependent cued memory was assessed (c). At phase 3—contextual memory, mice were exposed to the same context as day one for 5 min and contextual memory was assessed (d). Long-term spatial learning was assessed in the appetitive Y-maze, as determined by the percentage of times the mice made the correct choice as the first choice for reaching the goal (food reward) (e), as well as the number of average entries it took the mice to reach the goal (f). All data were normally distributed and analysed using a repeated measures ANOVA or one-way ANOVA, followed by a Dunnett’s post hoc test. Significant differences are depicted as *p < 0.05; milk control compared to kefir supplementation, ^$p < 0.05; undisturbed control compared to Milk control. All data are expressed as mean ± SEM (n = 10–12). Dots on each graph represent individual animals