Background

Zebrafish (Danio rerio) has emerged as a model organism for translational research in biomedicine, contributing to hundreds of studies on disease mechanisms, ranging from neurological and psychiatric disorders (Stewart et al., 2014), to cancer (White et al., 2013), and other age-associated diseases (Carneiro et al., 2016). Due to its robust phenotypes, and physiological (neuroanatomical, neuroendocrine, neurochemical) and genetic homologies to mammals, zebrafish has been considered as an ideal model for high-throughput behavioral assays for neurophenotyping and screening genetic mutations and psychotropic drugs. Here, we aim to describe a behavioral assay to measure judgment bias in zebrafish, in which some individuals consistently evaluate an ambiguous stimulus as negative (aka pessimists) and others as positive (aka optimists). Our assay is based on a judgment bias experimental paradigm originally developed for rodents (Harding et al., 2004), which was designed to measure expectations of positive (P) or negative (N) outcomes, when individuals are exposed to an ambiguous stimulus (A) that is intermediate between the two reference stimuli (P and N), and that they have learned to discriminate. After almost 20 years from its first publication, experimental tests to assess judgment bias have been developed and validated for a broad range of species (e.g., Mendl et al., 2010; Enkel et al., 2010; Bateson et al., 2011; Salmeto et al., 2011; Murphy et al., 2013). Most of the judgment bias studies have been focused on the effects of specific conditions inherent to life in captivity that are expected to impact the animal’s affective states. For instance, the effects of providing environmental enrichment (Douglas et al., 2012; Bethell and Koyama, 2015) and common handling procedures (Neave et al., 2013) have been broadly studied. Notably, the majority of judgment bias tasks have been shown to be very sensitive to manipulations that induce negative states (Baciadonna and McElligott, 2015). Furthermore, this behavioral paradigm has been used as a tool for assessing affective states in translational studies (Boleij et al., 2012; Rygula et al., 2015), reflecting the importance of judgment bias tests for biomedical research. However, the zebrafish have been unable to successfully perform judgment bias tasks, in large part due to their difficulties in categorizing more than one stimulus simultaneously. In fact, a number of different experimental designs based on Go/Go or Go/No-go tasks have already been tested in this species, in which the difficulties involved in selecting appropriate stimuli, balanced reward and punishment, and, importantly, in the simultaneous categorization of more than one stimulus were highlighted (Tan, 2017). Consequently, none of these assays was sufficiently reliable for judgment bias testing. In order to overcome these issues, we have used a combination of different stimulus classes (i.e., spatial and visual cues), which has been successfully operated in bumblebees (Perry et al., 2016) and may facilitate the acquisition of the relevant information for the performance of the task.

We believe the implications of the protocol reported here are of great significance for a broad variety of research fields due to its high applicability across research topics. The emerging evidence in the literature indicates that cognitive bias, and specifically judgment bias tasks, show promise as new measures of affective states in animals. Given the inaccessibility of direct reports of emotions in animals, an accurate assessment of their affective states is an important goal in animal welfare science. We also believe that this protocol is of general interest for translational scientists, with special mention of the research in psychiatric disorders, cancer, and other age-associated diseases, since it may provide an insight for inter-individual variation in susceptibility/resilience to these diseases. Considering the results of our Supporting Primary Paper (Espigares et al., 2021), we also believe that this protocol has the potential to be useful in the study of the in vivo efficacy of psychotropic drugs that have already been shown to exert a counteractive effect on telomere shortening. This is an important issue, since the correlation between the effect of telomerase activity by psychotropic medications (e.g., lithium and antidepressants) and their clinical efficacy still requires further investigation. On the other hand, a large number of psychological and neurophysiological studies suggest that judgments of ambiguous stimuli are the end result of a decision-making process that involves several components: sensory registration of the stimulus; evaluation of the stimulus and likely decision outcome; selection of a response. Therefore, the study of the cognitive and neural processes underlying judgment biases may also constitute a target for behavioral neuroscientists. Finally, our behavioral assay may have important applications in aquaculture research, since it could be easily implemented in other fish species with economic interest (e.g., sea bass, sea bream, salmon) by adapting the size of the behavioral apparatus.