Effects of repeated ayahuasca administration on behaviour and c-Fos expression in male rats exposed to the open field

Beyond ANOVA and MANOVA for repeated measures: Advantages of generalized estimated equations and generalized linear mixed models and its use in neuroscience research

“…Such advantages can be particularly useful in neuroscience research. For example, in recent works from our laboratory, we observed that the Normal distribution (positive, bell‐shaped, and symmetric around the mean) was not the most adequate for our fear conditioning, inhibitory avoidance, and open‐field data (Daldegan‐Bueno et al, 2022; Melo et al, 2020; Melo et al, 2021; Santos et al, 2020). Comparing different models using the QIC estimator (Vuong, 1989; Wedderburn, 1974), we observe that our freezing, latency, and locomotion measures fitted better to the Gamma probability distribution (positive, asymmetric, and skewed towards positive values).…”

“…When considering neuroscience studies, an adequate link function, as much as other analysis parameters (e.g., model and distribution), will depend on the interpretability of the results more than the model quality measures like QIC to evaluate the best analysis plan (Yu et al, 2022). For instance, in designs like ours (Daldegan-Bueno et al, 2022;Melo et al, 2020Melo et al, , 2021Santos et al, 2020) involving continuous measures (e.g., freezing, latency, and locomotion), the Identity link function seems to be an adequate choice because it assumes an additive relationship between dependent and independent variables, allowing a more direct, accurate and parsimonious interpretation of coefficients using betas (β) as differences of means between groups (Lo & Andrews, 2015).…”

“…Longitudinal designs enable individuals' investigation over time and under different conditions (Ballinger, 2004). For instance, in neuroscience research of learning and memory, we can observe repeated behaviours (e.g., locomotion and freezing or latency times) across successive time points (minute to minute) after or before conditioning sessions (session factor) and with or without drug effects (group factor) (Daldegan-Bueno et al, 2022;Melo et al, 2020Melo et al, , 2021Santos et al, 2020). Since these observations are from the same subjects, they may be correlated and requires a model that assumes dependence between observations (Liang & Zeger, 1986).…”

“…We used the GEE to analyse longitudinal data of the neural circuitry involved in learning and memory using fear conditioning, inhibitory avoidance, and open-field tasks in our laboratory (Daldegan-Bueno et al, 2022;Melo et al, 2020Melo et al, , 2021Santos et al, 2020). The GEE is recommended when a random effect is not necessary (homoscedastic model), whereas the GLMM is an approach recommended when a random effect is relevant for the study due to data heteroscedasticity or high interindividual variability (Muth et al, 2016).…”

Molecular Pathways of the Therapeutic Effects of Ayahuasca, a Botanical Psychedelic and Potential Rapid-Acting Antidepressant

“…In the last several decades, scientific research on ayahuasca consumption and administration has been conducted to investigate its possible therapeutic effects. From pre-clinical investigations [ 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 ] to observational studies [ 13 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 ], clinical trials [ 11 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ], reviews, and meta-analyses [ 2 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 ], there is accumulating evidence that ayahuasca and its isolated substances may present neuroprotective, neuroplastic, immunomodulatory, anticancer, anti-addictive, anxiolytic, and, most notably, antidepressant properties, while also increasing well-being and quality of life and improving emotion regulation. Investigations on the therapeutic properties of psychedelics are exploring possibilities to use these substances for treating several psychiatric disorders, especially depression [ 43 , 46 , 51 ].…”