Research sense: Incorporating animals’ sensory capacities in animal care and study design

Introduction

Considering animals’ sensory capacities is important in experimental design ¹ and for welfare reasons, as it can be difficult to predict the consequences of the ways animals experience the world. In rodents, for example, cage cleaning and ventilation can result in a variety of unexpected negative welfare effects.^2–5 In other species, differences in flicker fusion frequencies— the frequency at which a flashing light comes to be perceived as steady—can cause health problems. ⁶ These issues can also impede scientific validity, as there are important relationships between the 3Rs of animal welfare ⁷ and the recently added 3Vs (construct validity, internal validity, and external validity). ⁸ These problems may be even more common for non-model organisms, where sensory capacities may be unknown, standardized care practices do not exist, and peer reviewers and animal care experts are not well-positioned to assess relevant methods in ethics applications or manuscripts.

A better understanding of animals’ sensory capacities can, therefore, enhance both validity and welfare. Yet neither journals (the traditional guardians of validity) nor ethics committees (the traditional guardians of ethics) typically have procedural strategies to address sensory perception differences. It may therefore be wise to incorporate reflection on differences in animal sensory capacities during the experimental design and ethical approval processes.

Accordingly, we consider the pros and cons of three strategies that the scientific community could pursue to ensure that sensory differences are systematically considered in experimental review. While we focus on the experimental approval process here, these considerations could apply equally to setting up new animal facilities and the introduction of a new species to an existing facility. In exploring these pros and cons we hope to encourage a conversation and future research that can both benefit animal welfare and promote reproducible science.

The strategies

The first strategy we consider is the incorporation of specialist expertise (referred to as “expert consultation” or EC in Figure 1). Outside experts with a particular awareness of sensory modalities of studied organisms could be added to project teams. This is a common approach in the case of statistics, for example, where many universities offer consulting to their scientists through their statistics departments.

OPEN IN VIEWER

Figure 1.

Comparison of three proposed approaches along three key traits (using subjective assessments of the authors; no quantitative comparison is implied). The x axis represents the extent to which the approach incorporates expert review of factors related to sensory capacities. The y axis indicates the extent to which the approach is scientist-directed versus instigated by institutions. The color and size indicate the extent to which the approach is voluntary for scientists to participate in, as compared with required.

In the case of sensory experience, animal welfare scientists and sensory biologists might be the relevant experts upon whom to call for a “sensory review.” Universities with relevant departments in-house could offer similar internal consulting services. Alternatively, societies for the relevant disciplines could call upon their members to volunteer their services from time to time as a type of service obligation. Or, scientists could put consulting budgets into their grant applications to secure the paid services of such experts.

Because this approach reduces the investigative burden on the researcher, it is more likely to promote the intended objective when used. However, the comparison with statistics is instructive. Despite the availability of statistical consulting services, many scientists still make significant statistical mistakes at the experimental design stage and beyond.^9–12 In part, this might result from statistics experts not having capacity to offer support to biologists. If right, the statistical analogy offers two potential downsides of the approach, which are that there may not be a sufficient number of sensory biologists to provide this sort of consulting, or access to such experts might vary by institution or depend on the scientist’s personal networks. If so, growth of the sensory biology community could eventually make a more formalized version of this approach more feasible.

A second option might be voluntary disclosures or certification schemes (CS in Figure 1), where papers could include statements on how the authors considered the sensory capacities of their study animals, potentially referencing protocols developed by external bodies. The success of such an approach would depend on the ability of relevant experts to develop such protocols and the willingness of researchers to adopt them. Without a mandate, researchers might default to established procedures rather than investing in improvements. These challenges can be overcome with time—which could be reduced if journals require relevant statements. However, the efficacy of this strategy is questionable: data availability statements, for instance, largely have not resulted in changes in actual data availability in some contexts. ¹³

Scientists’ general reluctance to change their practices may be compounded by a legitimate interest in being able to compare new results with previous studies, which can create a kind of procedural inertia where established methods, even if suboptimal from the perspectives of both welfare and validity, persist simply because they are familiar (see, for example, the slow uptake of “tunnel” handling procedures in mice).^14,15 A third option, then, would be to overcome this conservatism via institutional requirements, such as instituting mandatory sensory review as part of the care and use or ethics review (ER in Figure 1). A simple “sensory modality survey” (see the Supplementary material online) could be added to the proposal process. The value of such a form depends on review committees guiding researchers and enforcing use. Researchers should be encouraged to consider both intended effects, such as the impact of a specific lighting schedule on circadian rhythms in a study on sleep, and indirect effects that might not be immediately apparent. A thorough review of relevant literature might often be warranted, with the side benefit that such a review could be published as a resource to be used by future scientists studying the same species. Such reviews could be useful even beyond academia, such as in zoos or rehabilitation centers. There are, therefore, good reasons to mandate the completion of a sensory modality survey (Supplementary material).

However, we anticipate that many animal ethics committees will be averse to adding an additional administrative burden to research. One reason for opposition could be fears of checklist proliferation and, second, fears of duplication of effort. In the first case, new guidelines have been suggested in recent years to improve scientific quality. Examples include the Sex Inclusive Research Framework ¹⁶ and the Replacement Checklist. ¹⁷ While these checklists advance laudable goals, it could be that a proliferation of such approaches becomes overly burdensome. A second reason for resistance is that a sensory modality survey may be seen as duplicating information already present elsewhere. For example, in the European Union (EU), EU Directive 2010/63/EU requires a severity assessment from the animal’s point of view, and a sensory modality survey might overlap with this.

To address these concerns, it may be possible to replace elements of existing ethical approval frameworks with a sensory approach, in cases where a sensory approach is shown to be more effective at eliciting a thorough and accurate listing of considerations. In the US, Institutional Animal Care and Use committees have broad leeway on the design of their approval forms, and so could experiment with modifying aspects of their documentation to incorporate a sensory perspective. If a sensory approach is taken instead of less effective approaches to documenting welfare impacts, there could be improvements for both animal welfare and efficiency of the process.

Future work

We have proposed three potential avenues for the incorporation of consideration of sensory experience into experimental design and ethical review (see Figure 1 for a comparison along several key traits). We hope that others will continue to propose alternative strategies. As they do, all approaches should be refined and compared to better understand what maximizes animal welfare and scientific validity while minimizing bureaucracy and cost. Controlled trials could be conducted comparing the animal welfare outcomes in projects that incorporate a given tool or approach to those that do not.

The strategies we have offered here could help identify and address potential welfare concerns while also controlling for disturbances that might affect research outcomes. While implementing any such strategy will require additional effort, we submit that the potential benefits for both animal welfare and scientific quality justify some investment.