Aims Certain bacteria can produce gamma aminobutyric acid (GABA) from glutamate in the human intestinal tract, leading to the possibility of altering GABA levels through diet. To this end, we assessed the ability of seven commercially available probiotic supplements to produce GABA. Method and results Probiotic strains were compared for GABA production in pure culture. The bacteria were inoculated at a concentration of 107 CFU ml−1 in 10 ml MRS supplemented with monosodium glutamate (1% w/v), both with and without oligofructose-enriched inulin (OFI) (1% w/v). Two strains with the highest production of GABA were further assessed for 48 h in pH-controlled anaerobic batch cultures inoculated with faecal bacteria. Liquid chromatography-mass spectrometry (LC–MS) was used for quantification of GABA and microbiota composition was determined through 16S rRNA gene sequencing. Levilactobacillus brevis LB01 (CGMCC 16921) and Lactiplantibacillus plantarum 299v (DSM 9843) were the most efficient producers of GABA. High GABA levels (28.32 mmol l−1 ± 0.29) were produced by the probiotic strain L. brevis LB01 at pH 5.4–5.6. This was significantly higher than the levels of GABA produced by L. plantarum (4.8 mmol l−1 ± 6.8) and a negative control (2.9 mM ± 3.1). The addition of OFI did not further stimulate GABA production under the conditions tested. The ability of these strains to produce GABA in-vitro was further evaluated in a faecal microbiota environment. Once again, L.brevis LB01 produced the highest levels of GABA (40.24 mmol l−1 ± 20.98). Conclusions L. brevis LB01 was found to be the most efficient probiotic strain, of those tested, for GABA production.

Previous research has demonstrated that individuals with high levels of Intolerance of Uncertainty (IU) have difficulty updating threat associations to safety associations. Notably, prior research has focused on measuring IU-related differences in threat and safety learning using arousal-based measures such as skin conductance response. Here we assessed whether IU-related differences in threat and safety learning could be captured using eye-tracking metrics linked with gaze behaviours such as dwelling and scanning. Participants (N = 144) completed self-report questionnaires assessing levels of IU and trait anxiety. Eye movements were then recorded during each conditioning phase: acquisition, extinction learning, and extinction retention. Fixation count and fixation duration served as indices of conditioned responding. Patterns of threat and safety learning typically reported for physiology and self-report were observed for the fixation count and fixation duration metrics during the acquisition and extinction phases. However, we failed to replicate IU-related patterns of disrupted safety learning that are typically observed with arousal-based measures for the eye-tracking metrics (e.g., greater differential responses during extinction learning and retention). Instead, we found that higher IU, while controlling for trait anxiety, was associated with reduced fixation duration (i.e., greater scanning behaviour) overall during extinction learning and retention. Our results inform models of anxiety, particularly in relation to how individual differences in IU modulate gaze behaviour during threat and safety learning.

Previous research has demonstrated that individuals with high levels of Intolerance of Uncertainty (IU) have difficulty updating threat associations to safety associations. Notably, prior research has focused on measuring IU-related differences in threat and safety learning using arousal-based measures such as skin conductance response. Here we assessed whether IU-related differences in threat and safety learning could be captured using eye-tracking metrics linked with gaze behaviours such as dwelling and scanning. Participants (N = 144) completed self-report questionnaires assessing levels of IU and trait anxiety. Eye movements were then recorded during each conditioning phase: acquisition, extinction learning, and extinction retention. Fixation count and fixation duration served as indices of conditioned responding. Patterns of threat and safety learning typically reported for physiology and self-report were observed for the fixation count and fixation duration metrics during acquisition and to some extent in extinction learning, but not for extinction retention. There was little evidence for specific associations between IU and disrupted safety learning (e.g., greater differential responses to the threat vs. safe cues during extinction learning and retention). While there was tentative evidence that IU was associated with shorter fixation durations (e.g., scanning) to threat vs. safe cues during extinction retention, this effect did not remain after controlling for trait anxiety. IU and trait anxiety similarly predicted greater fixation count and shorter fixation durations overall during extinction learning, and greater fixation count overall during extinction retention. IU further predicted shorter fixation durations overall during extinction retention. However, the only IU-based effect that remained significant after controlling for trait anxiety was that of fixation duration overall during threat extinction learning. Our results inform models of anxiety, particularly in relation to how individual differences modulate gaze behaviour during threat conditioning.