This study aims to establish a protocol for evaluating the object recognition memory and object location tasks in zebrafish. We evaluated novel the object recognition memory and analyzed the exploration time of the objects during training and testing. Zebrafish explored more the new object in comparison to the familiar object (61% of exploration time during test session). We also tested the object location task and measured the exploration time of each object in the familiar and novel object location. There was a preference to explore the object in the novel location (63% of exploration time during test session). The effect of the non-competitive NMDA receptor antagonist MK-801 was investigated on the object recognition and object location memory. Control (water only) and treated animals (5 μM MK-801) presented a significant preference in exploring the familiar object in comparison to the new object (66 and 68% of exploration time, respectively, during test session); however, 10 μM MK-801-treated animals did not show differences in the exploration time of the objects. In the object location task, the animals treated with the 5 or 10 μM MK-801 did not show a preference for the familiar or novel location whereas the control group had a higher preference in exploring the object in the familiar location (64% of exploration time during test session). Considering the different responses of the control group between original task and in the regimen treatment, we evaluated the impact of habituation on cortisol levels of animals in three different protocols: (1) habituated at the experiment apparatus for 3 days (C1 condition), (2) habituated at the experiment apparatus for 3 days plus treatment tank exposure at fourth day (C2 condition), (3) habituated at the treatment tank and experiment apparatus for 3 days and exposed to treatment tank again at fourth day (C3 condition). The results showed higher levels of cortisol in animals submitted to C2 and C3 conditions compared to animals submitted to C1. When introduced to an acute stressor during C1 condition, we observed an increase in the cortisol levels and an absence of preference for the objects in comparison to control group, which had a preference for novel object and novel location. Fluoxetine treatment induced a decrease in cortisol levels and an absence of preference for the objects in C2 and C3 conditions in comparison to control group, which had a preference for familiar object. However, fluoxetine treatment induced a preference to the novel location in C2 and C3 conditions in comparison to control group, which had a preference for familiar location. These results indicate that treatment tank exposure induced a different performance in object recognition and object location memory due to stress responses. Therefore, these tasks are prone to evaluate memory in physiological and pathological conditions, but its use is limited due to sensitivity to stress caused by manipulation.