To our knowledge, this is the first study to collect information on strepsirrhine primate training programs in North American institutions. The high level of survey completion rate of the participant institutions (between 96–100%) could represent the interest in the topic as other surveys with similar formats received lower completion rates [
54,
55,
56]. In fact, around 25 multiple choice questions could have potentially been too many for busy zoo staff members to answer completely [
54]. However, the high survey completion rate reflects the great effort and interest from the participating institutions. It is possible that explaining how the online survey works, distributing the questions in a pdf format ahead of time, grouping the questions clearly in three sections, and distribution through the AZA PTAG decreased factors that are commonly cited to affect response rates, including having a sponsoring organization, question order, question display, contact delivery modes, and pre-notification [
58,
59]. Although we covered only 57% of the AZA institutions that hold strepsirrhine species, the average response rates of online surveys in 2008 were calculated to be around 11% to 15% [
60], and it has been suggested to have dropped even further since then [
58]. Again, this could represent the attractiveness of the topic, and, in the future, it would be illuminating to expand the survey to other regions that hold these species. In particular, it would be interesting to survey facilities in Europe due to the high number of zoos and aquariums that manage these species, including eight species not housed in North America [
53].
4.1. Training Program Details, Animals, and Techniques
The high percentage of participant institutions that train strepsirrhine primates (97%) for husbandry and veterinary care (around 80%) suggests that these techniques are mainly being applied to enhance welfare through the cooperation of animals in many procedures, as cited previously in many other nonhuman primate species [
4,
5,
11]. In fact, with 92% of facilities doing scale training, 88% kennel training, and 78% separation training, with an average time in most of them (85%) of less than 30 min training strepsirrhines per day by individual staff members and 89% reporting training sessions under 10 min, the results based on staff responses reveal that behavioral goals can be achieved with relatively short time investment. Moreover, advanced behaviors, such as syringe training (45%) and injection training (38%), are emerging in multiple training programs, revealing that, although strepsirrhine training has not advanced to the same complexity as with larger nonhuman primates, the interest and skills to do so are making great progress. With 7% of facilities training for blood draw and 14% training for voluntary restraint, our survey reveals the development and trajectory for expanded opportunities that training strepsirrhines for voluntary participation is having in the behavioral management of these species.
Other benefits reported from these programs include increases in positive human–animal interactions, psychological well-being, and staff awareness of animal behaviors, as previously stated for other species [
2,
12,
13]. All these benefits, together with most organizations not reporting any risk in training strepsirrhine species (68%), confirm the success of these programs (as perceived by the staff) in contributing to the optimal care given to these nonhuman primates. It is also remarkable that no institution found monetary expenses as a concern, which is in contrast to what was reported for training programs for nonhuman primates in laboratories [
57]. Moreover, finding that 75% of institutions have been actively training strepsirrhine primates for more than five years (with 29% of them training them for more than 10 years) highlights the expansion that these techniques are having in zoos and research centers. These results also confirm a change of tendency from what was reported for North American zoos and related facilities in 2010, where the vast majority of institutions rarely or never conducted training with lorisid primates [
61]. In our survey, 15 institutions trained lorisid primates with Pygmy slow loris (
Nycticebus pygmaeus), the most commonly trained. Given that 20 facilities in North America house pygmy slow loris [
62], our survey indicates that a minimum of 55% of institutions with pygmy slow lorises in North America participate in a training program. Similarly, 86% of institutions that house aye-ayes (
Daubentonia madagascariensis) in North America [
63] participate in a training program, with six of seven institutions indicating the species as one participating in their training program.
It was not surprising to find that ring-tailed lemurs (
Lemur catta) were the species most commonly trained. Currently, 101 AZA facilities house ring-tailed lemurs [
64], making it the most numerous species of any strepsirrhine primate in North America. This diurnal species, with others such as ruffed lemurs (genus
Varecia) and some
Eulemur species, are highly social and live in multi-male/multi-female social groups of about 6–30 individuals [
65]. These species are usually maintained in larger groups and/or with mixed species, and behavioral management can be challenging, especially during breeding seasons, as reflected in our survey results. However, PRT has been documented as a useful tool to assist with behavioral management during challenging situations. Voluntary separations have been documented to eliminate behavioral indicators of fear or anxiety and teach individuals to remain calm for the duration of the separation [
43,
51].
4.2. Evaluation and Impact of Training
All the institutions, except three, document their training in some way, and the majority of them (70–78%) ensure consistency through meetings or sharing recorded data. With a growing understanding of the role of training in promoting opportunities for choice and positive engagement, documentation is important for program assessment and evidence-based decisions, as previously stated [
57,
66]. However, the lack of a unique method to record it or a unique platform, as evidenced by our survey, can make the evaluation of training programs difficult. In fact, only a small percentage of participants (19%) reviewed data from training sessions. There may be several reasons for this, with the main factor, as assumed by the authors, being the lack of time. Transcribing data from paper records to a computer-based analysis platform (e.g., Excel, ZIMS Care and Welfare, statistical or behavior analysis program) can be extremely time-consuming depending on the size of the training program, both in the number of trainers and/or individuals being trained. Records maintained in a digital database such as Excel will make analysis easier if in-house expertise is present to manipulate data and create graphs. While data analysis from training sessions can be challenging, progress is being made for user-friendly platforms that automatically create descriptive stats. Currently, a variety of tools, such as training templates in ZIMS, TRACKS, and other privately or individually developed software templates, are being developed and tested for the ease of data entry and output relevancy.
Another interesting result found in our survey was that although all the institutions used successive approximations to achieve behavioral goals, only 42% of them had criteria to determine if the animal has learned the shaping plan step. Again, we think that this can be related to the difficulties in analyzing training data and the lack of a unique system to record training sessions. However, adding the completion date of individual approximations as a trainer works toward the final goal combines two important elements of a training program, shaping plans and training records while also creating an additional output measurement to assess the total amount of time for an individual to learn a behavior and the duration of time to learn each step of the shaping process. Therefore, we recommend including it in the documentation because it can lead to the improvement of the analysis of training sessions, creating, along the way, opportunities to discuss training techniques and shaping plans. When discussing a challenging training situation, managers and trainers can look at the training data to identify when an animal’s learning has plateaued or regressed and adjust accordingly. Data analysis may also play an important role when assessing inter-species training goals and species-specific traits that contribute to or inhibit the successful completion of husbandry behaviors.
When asked about training differences, respondents found the personality of the individuals as one of the main factors (77%) that can influence training success, something that has also been previously suggested for many other nonhuman primate species [
17,
45,
46,
47,
67]. In fact, institutions in our survey also identified several personality traits as challenges when training strepsirrhine species, such as being “shy or skittish”, “hyperactive”, “hyper-reactive”, or “easily distracted”. However, the majority of them (94%) did not have a formal process to assess the personality of the individuals. Incorporating a formal process to evaluate personality, combining behavioral coding and trait rating methods [
44,
68], may not only provide a more thorough insight into these differences to adjust shaping plans, but it can also be applied in many areas to improve the welfare of nonhuman primates [
45,
68,
69,
70,
71]. Following these applications, several zoological associations, for example, the British and Irish Association of Zoos and Aquariums (BIAZA), have developed guidelines that include explanations of how to measure and analyze the personality and benefits that this research brings to different areas of health, management, and conservation of zoo animals [
72]. Additionally, there are books dedicated entirely to personality or temperament in nonhuman primates [
73], and, specifically for strepsirrhine species, several studies have measured it with behavioral coding identifying traits on a spectrum of boldness to shyness ([
74] with
Otolemur garnetti; [
75,
76] with
Microcebus murinus), while others have combined behavioral coding and trait rating ([
77] with
Nycticebus pygmaeus).
Another factor found to influence training, supported by our survey results, was the species. For example, tamarins were reported to more rapidly approach trainers and learn behaviors than marmosets [
19], and squirrel monkeys (
Saimiri spp.) learned a simple “target” behavior (touching a stationary object when presented) significantly faster than owl monkeys (
Aotus spp.), although these species did not differ in the amount of time required to train subsequent behaviors [
23]. For strepsirrhines, a study developed at the Duke Lemur Center with the
Eulemur species by the first author (G.F.L.), supervised by the second author (M.H.D.), also found significant differences related to the species [
78]. This research on 14 lemurs was performed as part of a Ph.D. dissertation, which found that red-collared lemurs (
Eulemur collaris) approached trainers and progressed in training (following a target around a short, simple obstacle course to complete an S shape) more rapidly than the other study species (
Supplementary S2). Additionally, white-fronted brown lemurs (
Eulemur albifrons) took the longest to approach trainers and showed the slowest progress in training, but no sex differences or partner effects were found (
Supplementary S2). Other researchers have also reported training differences between lemur species [
79]. Ring-tailed lemurs (
Lemur catta) ceased responding earlier than brown lemurs (
Eulemur fulvus) during the first extinction phase (i.e., the response is lost over time when a reward is no longer provided) and attained higher response rates during subsequent reinforcement sessions. However, with only one subject representing each species, it would be premature to conclude that a species difference has been demonstrated in that study. Additional data on training variations in lemur species may allow us to see if these results remain and to test whether differences persist in longer studies that include the training of subsequent behaviors. Longer studies would also allow us to test whether or not there is species variation in persistence to learn a new task, a trait that has been shown in red-fronted lemurs (
Eulemur rufriforns) to be important for individual success during innovation [
80], which could play an important role in training success.
Our survey results also revealed that although the application of training to promote positive animal welfare through voluntary participation is in line with best practices, as outlined in AZA guidelines and other nonhuman primate training programs [
4,
5,
11], most of the organizations (70%) did not monitor the impact of training on the well-being of these species. It is true that a positive indicator of staff perception of training programs with strepsirrhine species is that these programs contribute to the optimal care of the species; however, as Melfi and Ward stated [
81], the welfare impact of training has to be based on the empirical monitoring of its application. As an extension of the challenge of analyzing training records, the creation of assessment tools is an area of rapid expansion in the zoological community, as we have stated previously with personality. As animal welfare refers to the physical health, behavior, and emotional state of animals, measuring it requires a multi-dimensional approach combining different indicators [
82]. This can be complex, but we already have examples in strepsirrhine primate species that combine behavioral observations, personality ratings, and physiological measures [
77,
83]. With advanced husbandry behaviors increasing with strepsirrhine species, evidence-based physiological measures may become more prevalent. Voluntarily collected biological samples, such as blood, urine, and saliva, would help validate species-specific behavioral indicators currently used to describe good welfare and a positive affective state. The validation of the positive effects of choice, control, and engagement for an individual participating in a training program would benefit from a combination of both behavioral and physiological indicators. Utilizing a multi-dimensional approach to measure the efficacy of training to reduce negative stressors and increase positive effects could provide a deeper understanding of the impact of training on the welfare of strepsirrhine primates.