Can computers automate the assessment of fear responses in mice? This study introduces a computer-assisted method for quantifying freezing behavior in mice during Pavlovian fear conditioning, addressing the limitations of traditional human observation. The algorithm, using NIH Image on a Macintosh computer, simultaneously scores freezing in four mice with high accuracy. The computer scores correlated highly with human scores, accurately mimicking the effects of shock intensity and test modality on fear. Additional measures such as activity suppression and baseline activity scores are also acquired, enhancing the assay's sensitivity. This system allows for standardized and carefully controlled assessment of multiple aspects of the fear conditioning experience, which can detect interesting memory phenotypes and control for possible confounds. This method offers a reliable, efficient, and unbiased alternative for assessing fear conditioning, facilitating the screening of learning and memory deficits in mutant mice and advancing research in behavioral neuroscience.
This paper is a good fit for Learning & Memory, as it focuses on a novel method for assessing learning and memory processes. The study's investigation of fear conditioning in mice, a model relevant to understanding learning and memory mechanisms, is directly aligned with the journal's scope in neuroscience and behavioral research.