1 © 2022 Universities Federation for Animal Welfare The Old School, Brewhouse Hill, Wheathampstead, Hertfordshire AL4 8AN, UK www.ufaw.org.uk Animal Welfare 2022, 31: 1-12 ISSN 0962-7286 doi: 10.7120/09627286.31.1.001 Training as enrichment: A critical review EJ Fernandez School of Animal and Veterinary Sciences, University of Adelaide, Adelaide, SA 5005, Australia: email; edjfern@gmail.com; https://orcid.org/0000-0001-5444-6604 Abstract Husbandry training and environmental enrichment are both important advancements associated with current behavioural welfare practices. Additionally, the use of training procedures has been proposed as a form of enrichment, with the implication that training can produce beneficial behavioural welfare results. This paper examines the concept of training as enrichment through three distinct ways training procedures could enrich: (i) training facilitates enrichment usage; (ii) training modifies interactions, conspecific or otherwise; and (iii) training expands behavioural repertoires. Within each category, the paper focuses on past research that provides empirical support for training functioning as enrichment, as well as related areas of research that provide additional evidence. Previous studies support the claim that training is enriching, with additional research necessary to better understand how prevalent and under what conditions training procedures function as enrichment. Future training research should examine these potential enrichment effects, including methodology that allows for comparisons to traditional enrichment, the use of welfare diversity/variability indices, and the effects of learning on trainers and trainees alike. Keywords: animal welfare, enrichment, husbandry, learning, positive reinforcement, training Introduction The modern existence of animals under human care is connected to two major behavioural welfare advances: The use of animal training procedures to promote the husbandry of animals (Forthman & Ogden 1992; Desmond & Laule 1994; Laule et al 2003; Melfi et al 2020), and the imple- mentation of environmental enrichment (Markowitz 1982; Shepherdson et al 1998; Young 2003; Maple & Perdue 2013). Animal training can be defined by respondent and operant conditioning procedures used to elicit, evoke, or emit behaviour (Pryor 1999; Pierce & Cheney 2013; Domjan 2014; Pryor & Ramirez 2014; Ramirez 2020). For instance, in the case of an operant conditioning procedure, food or some other consequence is delivered as a reward for engaging in a selected response. It is also worth noting that most of the focus on modern animal training presented within this review emphasises positive reinforcement and similar force-free applications to effectively change behaviour. Environmental enrichment can be defined as stimuli and/or events that are added to or modify an animal’s environment and result in some measurable improvement in behavioural and/or physiological well- being/welfare (Newberry 1995; Shepherdson 1998; Mellen & MacPhee 2001; Fernandez & Timberlake 2008; Hoy et al 2010; Fernandez et al 2021a). Some examples of enrich- ment include the use of foraging devices and feeding schedules, both automated and non-automated (Carlstead et al 1991; Shepherdson et al 1993; Fernandez 2010; Andrews & Ha 2014; Bashaw et al 2016; Fernandez 2021), changes in enclosure presentations, including choice between enclosures (Carlstead et al 1993; Sherwin et al 1999; Coe 2004), and the presentation of auditory, olfactory, and/or visual stimuli (Carlstead & Seidensticker 1991; Platt & Novak 1997; Graham et al 2005; Wells & Irwin 2008; Fernandez & Timberlake 2019a). While both training and enrichment advances have remained relatively autonomous, the concept of training as a form of enrichment itself has been proposed (Laule & Desmond 1998; Laule 2003; Laule & Whittaker 2007; Brando 2012; Melfi 2013, 2014; Westlund 2014; Melfi & Ward 2020). The implication is that enrichment is a means to improve the welfare of captive animals, and training is proposed to improve welfare, and is therefore enriching. However, what remains less clear are the ways animal training procedures could be empirically measured to have an enriching effect. Some authors, such as Laule and Desmond (1998) and Westlund (2014) have proposed training enriches by providing animals with greater choices and control over their environment, although are less specific about how such choice and control could be demonstrated through an observable metric. Alternatively, Melfi (2013, 2014) proposed several directly testable Universities Federation for Animal Welfare Science in the Service of Animal Welfare