Effects of simple cage enrichment and its removal on the behavior and welfare of American mink (Neogale vison)

ABSTRACT

Environmental enrichment may reduce stereotypies in fur-farmed mink. North American mink standards require manipulable enrichment objects within cages. However, mink can rapidly destroy objects inhibiting continuous enrichment presence, which may have negative welfare impacts. This experimental study determined the effects of removing simple cage manipulable enrichments (plastic chains and dumbbells), either short-term or longer-term, on the behavioral expression of welfare in fur-farmed mink. Locomotor stereotypies, normal activity, sub-types of inactivity related to boredom, and tail fur-chewing were recorded across four treatment groups with either (1) no enrichment, (2) continuous enrichment, (3) short (temporary), or (4) long-term enrichment removal. Contrary to predictions, locomotor stereotypies, and scrabbling were not reduced by the enrichments, nor affected by the enrichment removal. Observations at the beginning of the trial showed the non-enriched mink spent the least amount of time lying with their eyes open (i.e., the least bored). The lack of enrichment may have increased fur-chewing on the tail, but larger sample sizes would be needed for statistical confirmation. This research contributes to the literature on evaluating simple, practical enrichments for improving fur-farmed mink welfare.

KEYWORDS:

• fur-chewing
• stereotypy
• boredom
• Neovison vison
• Mustela vison

Introduction

Globally, for countries where fur farming remains legal, millions of mink (American mink: Neogale vison, formerly Neovison vison, Mustela vison) are raised for their pelts. Farmed mink are typically housed (singly or in small groups) in wire cages of varying sizes (Díez-León et al., 2017; Henriksen et al., 2022) with different degrees of environmental complexity required by governing standards, guidelines, and welfare assessment protocols (e.g., Fur Commission USA, 2019; Møller et al., 2015; National Farm Animal Care Council [NFACC], 2021). Mink are semi-aquatic animals, and these smaller caged environments are disparate from their home ranges in the wild that can span multiple hectares in size, or linear kilometers along waterways (Ireland, 1990; Melero et al., 2008). Commercially, it is typically not feasible to provide swimming water to mink because of the logistical practicalities as well as health risks associated with water contamination, although some legislation mandates water provision (Schwarzer et al., 2016). Farmed mink frequently engage in varying negative or undesirable behaviors. This may include locomotor stereotypic behavior such as pacing, bobbing and scrabbling, fur-chewing, aggression, or high levels of fear (Hansen & Møller, 2001; Meagher et al., 2014) and signs of boredom such as inactivity (Meagher et al., 2012). These behaviors are deemed negative/undesirable as they are abnormal, injurious, or indicative of a negative affective state and can be symptomatic of a poorer welfare state (Meagher et al., 2013).

Experimental research with elaborate enrichment schemes that included running water access has demonstrated the positive impact of enrichment on reducing abnormal stereotypic behavior and fecal cortisol metabolites (Campbell et al., 2013; Dallaire et al., 2012; Díez-León et al., 2013, 2016; Meagher et al., 2013). Research with simple enrichment schemes, both in experimental settings and on commercial farms, has shown some effects of modifications such as wire shelves, cylinders, and manipulable objects/toys. These enrichments can reduce stereotypic behavior, tail-chewing, fecal corticoid metabolites, fearfulness, and aggressiveness, although not consistently across all sampled animals (Hansen et al., 2007; Meagher et al., 2014). There is also some evidence of enriched mink being graded as having better pelt quality than their non-enriched siblings and showing cleaner cages (Meagher et al., 2014). Given the evidenced benefits of manipulable enrichments, guidelines for farms within North America currently include the requirement for at least one simple manipulable cage object (by the end of 2013: National Farm Animal Care Council [NFACC], 2021, by the beginning of 2023: Fur Commission USA, 2019). However, providing manipulable enrichments to mink is challenging due to their jaw strength and ability to rapidly chew/destroy many objects. Practically, enrichments may be temporarily unavailable to the mink on-farm if the animal destroys their object (e.g., Malmkvist et al., 2013), or longer-term, if the enrichments need to be removed due to safety concerns during different on-farm stages such as whelping. Enrichments removed either temporarily or longer-term could create frustration in the animals, or alternatively, have a protective beneficial effect on the animal’s welfare in their absence.

Previous research with laboratory mice shows juvenile enrichment with adulthood removal can have two types of effects – “protection” or “frustration.” Captive-striped mice (Rhabdomys) provided with enrichments as juveniles showed reduced stereotypic behavior that was maintained when mice were transferred to standard cages in adulthood (Jones et al., 2011) – lending support for “protection.” However, another study using laboratory IRCD-1 mice instead found mice transferred from enriched to standard cages performed more stereotypic behavior than those raised in standard cages from birth – indicating “frustration” (Latham & Mason, 2010). When young adult mink (male and female) were moved from enriched cages (elaborate, included swimming water) to non-enriched cages, there were no detectable changes in total inactivity (Meagher et al., 2013). Females did show a decrease in time spent resting, and both males and females increased the time spent lying awake, increased the proportion of inactivity lying with their belly down, and decreased the proportion of inactivity curled up (Meagher et al., 2013). Transferring young adult mink from enriched (elaborate, included swimming water) to non-enriched cages had no impact on the already decreased locomotor stereotypic behavior in enriched mink, but within two weeks, enriched and non-enriched mink showed the same levels of stereotypic scrabbling (Díez-León et al., 2016).

Using simple additional cage items, the objectives of the present research were to (1) assess the impacts of practical, manipulable enrichment items including the effects of their short-term removal on mink behavioral expression of welfare, simulating enrichments being destroyed by mink with time lapses before replacement and (2) determine the effects of long-term enrichment removal on mink behavioral expression of welfare simulating movement of breeder mink to non-enriched cages after grading season. It was predicted that Group (1) (non-enriched) would display the most stereotypies, scrabbling, inactivity, and fur-chewing compared to mink in enriched cages with (Groups 3 and 4) and without enrichment removal (Group 2). Groups (3) (short-term enrichment removal) and (4) (long-term enrichment removal) were predicted to either increase stereotypy, scrabbling, inactivity, and fur-chewing levels after enrichments were removed – indicating frustration and boredom – or their levels would remain the same and be comparable to Group (2) (continuous enrichment), indicating a protective effect of having enrichment in the growth period. If removal resulted in increased undesirable behaviors, Group (4) was predicted to show greater increases relative to Group (3) due to the longer period of enrichment absence.

Methods

Ethical Statement

All the research was approved by the Michigan State University Institutional Animal Care and Use Committee (AUF #07/13–178–00).

Animals, Housing, and Treatments

We used 80 (40 female and 40 male) black color-type mink singly housed in one outdoor shed at Michigan State University Experimental Fur Farm, East Lansing, MI USA from July 2013 until February 2014. December through February were winter months and there were often freezing temperatures. Mink were exposed to natural light and ambient temperatures with a standard meat-based wet diet fed daily in the afternoon and ad libitum water access via nipple drinkers. All animals were visually checked daily to ensure optimal health and early detection of any issues. Over two consecutive days in late July, 2013, 20 kit groups of four littermates each were selected and placed into individual wire cages (62 cm L × 25 cm W × 38 cm H) with drop-in plastic and wire mesh nest boxes (24.5 cm L × 24 cm W x 29.5 cm H). Kits were approximately three months old having been born in April/May 2013. Single housing for juvenile and adult mink occurs on North American commercial farms (Fur Commission USA, 2019; National Farm Animal Care Council [NFACC], 2021). Littermates were placed in consecutive order and one of each of the four kits was randomly assigned to one of the four treatment groups (see below) to evenly distribute groups throughout the shed. The four littermates were either all male or all female to balance sexes across treatment groups. The sample sizes were based on similarity to previous studies that have detected significant behavioral differences (Campbell et al., 2013; Dallaire et al., 2012; Meagher et al., 2013) as well as within the limitations of available sex and sibling-matched mink kits from breeding females in the experimental setting. In early August 2013 one plastic dumbbell (Bio-Serv®, Flemington, NJ, USA) and a 25 cm length of plastic chain were added to 60 of the 80 cages (three kits from each group) as enrichment. The dumbbell was free-moving, while the plastic chain hung down from the cage wire ceiling in a fixed location. Any chains that were chewed and destroyed were immediately replaced. The dumbbells remained intact throughout the experiment. The treatment groups comprised varying durations of enrichment provision as follows:

Group 1: No enrichment for the duration of the experiment.

Group 2: Continuous enrichment for the duration of the experiment.

Group 3: Short-term enrichment removal: Enrichments provided for the duration of the experiment but temporarily removed for one week on two occasions (November 4–10, and December 8–14).

Group 4: Long-term enrichment removal: Enrichments provided until December 8 2013, then permanently removed until the experiment concluded February 16 2014.

Behavioral Observations

Commencing in late September of 2013, scanning live behavioral observations of all mink were conducted by a single experienced observer in the morning, prior to their normal afternoon feeding time, over a period of six days. The observations began seven weeks after enrichments had been provided to capture behavioral effects after a sustained period of exposure. The recorded behaviors (stereotypic, normal, and resting-related), including mink location (nest box or cage), are outlined in the ethogram presented in Table 1. The ethogram was based on previous studies of the same mink color-type in similar experimental facilities (Campbell et al., 2013, 2016; Meagher et al., 2013). These six observation days were consecutive where possible, but if there was a missed day for logistical reasons, observations occurred the following day to maintain the same number of observations across each six-day period. On each day, mink were scanned every 15 sec for 4 min, repeated twice per day. The observer stood quietly in front of the cages (approximately 1 m away) and observed four mink simultaneously for 4 min before moving to the next four mink. When all 80 mink had been observed once, the observations were repeated for a second time that day. Two sets of 4-min observations took approximately 2–3 h and occurred between approximately 08:30 until 12:00. Feeding occurred later in the afternoon when observations had finished. Following the first set of observations, behavioral data were again collected at the end of October (“October Prior”), after which enrichments were removed from Group 3 (short-term enrichment removal). Commencing the morning of the removal, another six days of observations took place (“November After” removal) and enrichments were then replaced at the end of the sixth observation day. The same procedure occurred in December where six days of observations took place (“December Prior”), then enrichments were removed from both Group 3 (short-term removal) and Group 4 (long-term removal) early December 2013, and another six days of observations took place (“December After”) commencing the morning of removal. The enrichments were then replaced for Group 3 only. Two final six-day observation periods were conducted in mid-January and mid-February 2014 after which the project concluded on February 16 2014, when the mink were approximately 10 months old. A timeline of the observation schedule is depicted in Figure 1.

Figure 1. The timeline depicting the experimental schedule from August 2013 to February 2014. Four treatment groups were provided enrichment for varying time periods including no enrichment (Group 1), continuous enrichment (Group 2), short-term removal of enrichment on two occasions (Group 3), and long-term enrichment removal (Group 4). Seven live observation periods (ob) are indicated including when tail-scoring (TS) occurred. The observation periods were classified as “September, Oct Prior, Nov After, Dec Prior, Dec After, Jan, Feb” based on month and whether they occurred just prior to or just after enrichments were removed in the short-term and/or long-term removal groups. Grading of the mink to assess suitable mink for the next breeding season is also indicated. December through February were winter months.

Table 1. An ethogram of the behaviors and postures recorded during live observations as well as the location of the mink.

Observed behavior and postures^a,b	Description
Normal activity	Walking, sniffing, eating, drinking
Scrabbling	Scratching with front paws at a surface
Locomotor stereotypies	3 or more repetitions of circling, pacing, bobbing, or pacing in and out of the nest box
Borderline stereotypies	Less than 3 repetitions of circling, pacing, bobbing, or pacing in and out of the nest box
Tail chasing	A form of stereotypy – repetitive chasing of the tail
Tail chewing	A form of abnormal behavior – repetitive chewing of the tail to the point of injury
Enrichment use (chain)	Touching and interacting with the plastic chain in the cage
Enrichment use (dumbbell)	Touching and interacting with the dumbbell in the cage or nest box (does not include sleeping next to the enrichment)
Grooming	Mink is licking their fur to groom themselves (distinct from tail chewing)
Sleeping	Resting, curled, with eyes closed
Resting	Resting, curled, with eyes open
Lying belly up, eyes open	Mink is lying on its back, not curled up, eyes are open
Lying belly up, eyes closed	Mink is lying on its back, not curled up, eyes are closed
Lying belly down, eyes open	Mink is lying on its belly, not curled up, eyes are open
Lying belly down, eyes closed	Mink is lying on its belly, not curled up, eyes are closed
Location
Home cage or nest box	Mink is present in either the cage or their nest box. (If the mink is route tracing in and out of the nest box, location was listed as “home cage”).

^aScrabbling was recorded separately from locomotor stereotypies as they likely have different underlying causes (e.g., Dallaire et al., 2011; Díez-León et al., 2016; Polanco et al., 2017).

^bDifferent types of inactivity were recorded as subtypes may have different welfare consequences (Meagher et al., 2013).

Tail Scoring

In addition to the behavioral observations, tails were visually scored on four occasions by a single observer for self-inflicted fur-chewing damage using a scoring system developed to estimate the depth of fur chewing (Table 2). The length of chewing damage was also visually estimated in centimeters. There was an opportunity for fur to grow back for an individual between some observation points when mink were in a pelt growing period (the exact timeline of pelt growth in these mink was not tracked within this study). Some previous research has indicated the majority of mink (~80%) in more barren conditions may exhibit tail-chewing (Hansen et al., 2007), whereas some commercial farm studies showed only up to ~10% occurrence (Meagher et al., 2014). It was uncertain how prevalent this behavior would be across the limited treatment sample sizes.

Table 2. The scoring system for describing the depth of tail fur-chewing.

Score	Description
1	Skin visible
2	Very short but skin not visible
3	Depth of fur is ¼ chewed
4	Depth of fur is ½ chewed
5	No fur-chewing

Data and Statistical Analyses

One female mink died prior to the final set of observations in February and thus only 79 mink were observed at the last time point. All analyses were conducted in JMP 17.0.0 (SAS Institute, Cary, NC, USA) with α < 0.05. Raw data are presented in the tables and figures with analyses conducted on transformed data where stated.

Use of the plastic chain or dumbbell were combined and calculated as a proportion of total normal activity (see Table 1, normal activity was combined with grooming). These data are visually presented but not statistically analyzed given the treatment interventions of removing the enrichments.

The behaviors and location (home cage or nest box) of the individual mink were converted to proportions of total number of observations across each six-day observation period so that there was a single proportional value per mink per observation period (80 mink x 7 observation periods minus one mink that died in the last observation period: n = 559) for each summarized behavior as follows:

• The proportion of total number of observations that were classified as activity (including all stereotypies but not resting/sleeping, which was inactivity; see Table 1 for ethogram).

• The proportion of the total number of observations classified as normal activity (including enrichment use and grooming, excluding all stereotypies, see Table 1 for ethogram).

• Stereotypies (locomotor, borderline, and tail chasing and abnormal behavior of tail-chewing, but excluding scrabbling) as a proportion of total activity (i.e., not resting/sleeping). Stereotypies (excluding scrabbling) were combined as there were few observations of borderline, tail-chasing, or tail-chewing behavior (117, 60, 71 observations, respectively, out of 121,138 recorded datapoints).

• Scrabbling as a proportion of total activity (assessed separately based on classifications in the literature: e.g., Campbell et al., 2013; Polanco et al., 2017; Polanco et al., 2018).

• The total proportion of inactivity (all subtypes of resting/sleeping, see Table 1 for ethogram) was the corresponding opposite to total activity, so only the total proportion of all lying down with eyes open was compiled for analysis (with one male outlier with a proportional value of one in the long-term removal group removed from the Jan observation period as he was deemed not representative. This value of one indicated that this male was only observed lying with his eyes open during this period, compared to the mean ± SEM of 0.06 ± 0.22 for the rest of the individuals in the long-term removal group).

• The proportions of occupancy in the two locations summed to one, thus analyses were conducted on the proportion of time spent in the home cage only.

The proportional behavioral and location data were logit-transformed with 0.0001 either added or subtracted for values of zero and one, respectively. Transformed data were analyzed using separate general linear mixed models by observation period with treatment group, sex, and their interaction as fixed effects and Family ID as a random effect. Restricted maximum likelihood estimation methods were applied. The studentized residuals were visually inspected for homoscedasticity. Where significant differences were present, Tukey’s HSD tests were applied post-hoc.

Tail score data across the four tail-scoring days were collated and displayed to show the frequency of chewing scores across the four treatment groups on the four days. No formal statistical analyses were conducted given the small sample sizes of mink that displayed tail fur-chewing.

Results

The mink were observed utilizing their enrichments across each observation period. On average, the proportions of enrichment use were typically less than 7% of their time spent engaging in normal activity (Figure 2). When visually compared with previous observation periods, there was lower enrichment use across the winter observation periods (December, January, and February) for the two groups with enrichment remaining, but this was not statistically analyzed (Figure 2).

Figure 2. The mean proportion (±SEM) of enrichment use as a proportion of total normal activity (excluding stereotypies) in mink from three treatment groups (enriched, short-term removal, and long-term removal) across the seven observation periods (September – February). “Prior” and “After” indicate observation periods prior to and after short-term enrichment removal (October/November and December) and long-term enrichment removal (Dec).

Overall, there were few, inconsistent, significant effects on the behavioral endpoints measured. There was no effect of the treatment group (all p ≥ .11) or sex (all p ≥ .13) on the proportion of total activity or proportion of normal activity displayed at each of the seven observation periods except for an effect of treatment on the proportion of total activity in the final Feb observation period (p =.02, Table 3). Post-hoc tests showed there was higher activity in the enriched treatment group compared with the short-term removal treatment group in this final observation period (Table 3). There was a significant interaction between treatment and sex (both p ≤ .04, all other p ≥ .06) for both the proportion of total activity and proportion of normal activity in the Dec After observation period only but post-hoc tests revealed no differentiation among groups (Table 3). There was no effect of the treatment group (all p ≥ .27) or sex (all p ≥ .20) on stereotypy performance as a proportion of total activity. There was a significant interaction between treatment and sex in the Feb observation period only (p =  .004, all other p ≥ .09) but post-hoc tests revealed no differentiation among groups (Table 3). Similarly, there was no effect of treatment group (all p ≥ .24) or sex (all p ≥ .06) on scrabbling as a proportion of total activity (Table 3). There was a significant interaction between treatment and sex in the Oct Prior observation period only (p = .04, all other p ≥ .06) but post-hoc tests revealed no differentiation among groups (Table 3).

Table 3. The statistics for each general linear mixed model that was conducted on the logit-transformed proportional behavioral data across 7 observation periods and mean (± SD) for the raw values for each behavioral variable per treatment group (enriched, non-enriched, ST (short-term) removal, LT (long-term) removal).

		Observation Period
Behavior	Treatment group	Sept	Oct Prior	Nov After	Dec Prior	Dec After	Jan	Feb
Proportion of total activity		Treatment: F(3,54) = 0.58, p = 0.63	Treatment: F(3,54) = 1.86, p = 0.15	Treatment: F(3,54) = 0.85, p = 0.48	Treatment: F(3,54) = 0.60, p = 0.62	Treatment: F(3,54) = 0.49, p = 0.69	Treatment: F(3,54) = 0.97, p = 0.42	Treatment: F(3,53.33) = 3.69 P  = 0.02
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 1.53, p = 0.23	F(1,18) = 2.46, p = 0.13	F(1,18) = 1.65, p = 0.21	F(1,18) = 0.48, p = 0.50	F(1,18) = 0.11, p = 0.75	F(1,18) = 1.30, p = 0.27	F(1,18.11) = 1.95, p = 0.18
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 0.45, p = 0.72	F(3,54) = 1.54, p = 0.22	F(3,54) = 1.40, p = 0.25	F(3,54) = 1.05, p = 0.38	F(3,54) = 5.21, P  = 0.003	F(3,54) = 2.63, p = 0.06	F(3,53.33) = 1.19, p = 0.32
	Non-enriched	0.32 ± 0.18	0.29 ± 0.18	0.35 ± 0.19	0.34 ± 0.22	0.29 ± 0.17	0.60 ± 0.23	0.64 ± 0.25
	Enriched	0.37 ± 0.14	0.27 ± 0.14	0.31 ± 0.18	0.36 ± 0.17	0.34 ± 0.17	0.70 ± 0.24	0.70 ± 0.25
	ST removal	0.35 ± 0.18	0.34 ± 0.17	0.34 ± 0.15	0.31 ± 0.19	0.34 ± 0.19	0.61 ± 0.23	0.53 ± 0.23
	LT removal	0.38 ± 0.17	0.29 ± 0.15	0.39 ± 0.21	0.33 ± 0.18	0.32 ± 0.18	0.67 ± 0.23	0.65 ± 0.21
Proportion of normal activity		Treatment: F(3,54) = 1.25, p = 0.30	Treatment: F(3,54) = 2.13, p = 0.11	Treatment: F(3,54) = 0.48, p = 0.70	Treatment: F(3,54) = 0.59, p = 0.63	Treatment: F(3,54) = 1.34, p = 0.27	Treatment: F(3,54) = 0.39, p = 0.76	Treatment: F(3,52) = 0.32, p = 0.81
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 1.15, p = 0.30	F(1,18) = 2.24, p = 0.15	F(1,18) = 0.67, p = 0.42	F(1,18) = 0.14, p = 0.72	F(1,18) = 0.0004, p = 0.98	F(1,18) = 0.07, p = 0.79	F(1,16.84) = 0.04, p = 0.83
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 0.75, p = 0.53	F(3,54) = 2.47 p = 0.07	F(3,54) = 1.23, p = 0.31	F(3,54) = 0.53, p = 0.66	F(3,54) = 2.98, P  = 0.04	F(3,54) = 0.55, p = 0.65	F(3,52) = 2.33, p = 0.08
	Non-enriched	0.22 ± 0.10	0.22 ± 0.12	0.27 ± 0.14	0.26 ± 0.13	0.20 ± 0.11	0.37 ± 0.11	0.33 ± 0.15
	Enriched	0.30 ± 0.12	0.22 ± 0.11	0.25 ± 0.13	0.26 ± 0.13	0.27 ± 0.13	0.41 ± 0.17	0.33 ± 0.14
	ST removal	0.26 ± 0.14	0.28 ± 0.11	0.27 ± 0.11	0.24 ± 0.13	0.23 ± 0.12	0.40 ± 0.14	0.32 ± 0.10
	LT removal	0.27 ± 0.09	0.21 ± 0.08	0.26 ± 0.11	0.24 ± 0.09	0.22 ± 0.10	0.40 ± 0.16	0.35 ± 0.16
Stereotypies as a proportion of total activity		Treatment: F(3,54) = 0.06, p = 0.98	Treatment: F(3,54) = 0.89, p = 0.45	Treatment: F(3,54) = 1.33, p = 0.27	Treatment: F(3,54) = 0.18, p = 0.91	Treatment: F(3,54) = 0.42, p = 0.74	Treatment: F(3,54) = 1.14, p = 0.94	Treatment: F(3,53.17) = 0.36, p = 0.78
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 0.59, p = 0.45	F(1,18) = 0.34, p = 0.57	F(1,18) = 1.09, p = 0.31	F(1,18) = 1.75, p = 0.20	F(1,18) = 0.02, p = 0.89	F(1,18) = 1.34, p = 0.26	F(1,18.01) = 0.35, p = 0.56
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 0.54, p = 0.66	F(3,54) = 0.95, p = 0.42	F(3,54) = 1.05, p = 0.38	F(3,54) = 0.84, p = 0.48	F(3,54) = 2.30, p = 0.09	F(3,54) = 2.32, p = 0.09	F(3,53.17) = 4.98, P  = 0.004
	Non-enriched	0.19 ± 0.21	0.13 ± 0.22	0.17 ± 0.20	0.14 ± 0.17	0.19 ± 0.23	0.31 ± 0.21	0.39 ± 0.29
	Enriched	0.13 ± 0.13	0.09 ± 0.15	0.10 ± 0.17	0.19 ± 0.24	0.14 ± 0.17	0.33 ± 0.28	0.43 ± 0.31
	ST removal	0.19 ± 0.22	0.11 ± 0.14	0.15 ± 0.15	0.16 ± 0.18	0.22 ± 0.23	0.28 ± 0.24	0.32 ± 0.26
	LT removal	0.22 ± 0.22	0.17 ± 0.19	0.22 ± 0.23	0.16 ± 0.20	0.21 ± 0.24	0.32 ± 0.29	0.36 ± 0.33
Scrabbling as a proportion of total activity		Treatment: F(3,54) = 0.21, p = 0.89	Treatment: F(3,54) = 1.45, p = 0.24	Treatment: F(3,54) = 0.90, p = 0.45	Treatment: F(3,54) = 0.17, p = 0.92	Treatment: F(3,54) = 0.72, p = 0.54	Treatment: F(3,54) = 1.17, p = 0.33	Treatment: F(3,53.76) = 0.59, p = 0.62
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 0.13, p = 0.72	F(1,18) = 0.22, p = 0.64	F(1,18) = 3.91, p = 0.06	F(1,18) = 1.30, p = 0.27	F(1,18) = 0.95, p = 0.34	F(1,18) = 0.19, p = 0.66	F(1,18.28) = 0.01, p = 0.90
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 1.30, p = 0.28	F(3,54) = 2.96, P  = 0.04	F(3,54) = 1.39, p = 0.26	F(3,54) = 0.33, p = 0.80	F(3,54) = 1.68, p = 0.18	F(3,54) = 2.65, p = 0.06	F(3,53.76) = 0.37, p = 0.77
	Non-enriched	0.04 ± 0.12	0.03 ± 0.07	0.005 ± 0.02	0.02 ± 0.05	0.03 ± 0.07	0.02 ± 0.05	0.02 ± 0.05
	Enriched	0.04 ± 0.08	0.05 ± 0.09	0.02 ± 0.06	0.02 ± 0.04	0.02 ± 0.06	0.02 ± 0.04	0.02 ± 0.07
	ST removal	0.01 ± 0.03	0.03 ± 0.07	0.03 ± 0.07	0.02 ± 0.05	0.009 ± 0.02	0.002 ± 0.004	0.009 ± 0.03
	LT removal	0.01 ± 0.03	0.01 ± 0.03	0.02 ± 0.06	0.02 ± 0.07	0.009 ± 0.03	0.006 ± 0.01	0.005 ± 0.02
Proportion of all lying with eyes open		Treatment: F(3,54) = 7.48, P  = 0.0003	Treatment: F(3,54) = 1.12, p = 0.35	Treatment: F(3,54) = 0.65, p = 0.59	Treatment: F(3,54) = 0.53, p = 0.66	Treatment: F(3,54) = 0.46, p = 0.71	Treatment: F(3,52.65) = 2.24, p = 0.09	Treatment: F(3,53.51) = 0.84, p = 0.48
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 2.69, p = 0.12	F(1,18) = 5.23, P  = 0.03	F(1,18) = 2.85, p = 0.11	F(1,18) = 0.08, p = 0.78	F(1,18) = 0.06, p = 0.81	F(1,17.35) = 1.54, p = 0.23	F(1,18.11) = 0.10, p = 0.75
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 2.94, P  = 0.04	F(3,54) = 1.04, p = 0.38	F(3,54) = 1.05, p = 0.38	F(3,54) = 1.02, p = 0.39	F(3,54) = 0.83, p = 0.48	F(3,52.65) = 0.70, p = 0.56	F(3,53.51) = 2.12, p = 0.11
	Non-enriched	0.05 ± 0.03	0.05 ± 0.05	0.02 ± 0.02	0.006 ± 0.01	0.002 ± 0.006	0.02 ± 0.04	0.01 ± 0.03
	Enriched	0.10 ± 0.08	0.07 ± 0.06	0.03 ± 0.03	0.01 ± 0.02	0.004 ± 0.007	0.009 ± 0.02	0.007 ± 0.02
	ST removal	0.12 ± 0.08	0.08 ± 0.06	0.05 ± 0.05	0.02 ± 0.03	0.005 ± 0.01	0.03 ± 0.07	0.002 ± 0.004
	LT removal	0.10 ± 0.07	0.05 ± 0.06	0.03 ± 0.04	0.01 ± 0.02	0.003 ± 0.006	0.06 ± 0.22	0.002 ± 0.008
Proportion of time in the home cage		Treatment: F(3,54) = 1.08, p = 0.36	Treatment: F(3,54) = 0.46, p = 0.71	Treatment: F(3,54) = 1.52, p = 0.22	Treatment: F(3,54) = 1.03, p = 0.39	Treatment: F(3,54) = 0.53, p = 0.67	Treatment: F(3,54) = 0.58, p = 0.63	Treatment: F(3,53.31) = 2.54, p = 0.07
		Sex:	Sex:	Sex:	Sex:	Sex:	Sex:	Sex:
		F(1,18) = 0.62, p = 0.44	F(1,18) = 0.50, p = 0.49	F(1,18) = 0.02, p = 0.90	F(1,18) = 0.67, p = 0.42	F(1,18) = 0.36, p = 0.55	F(1,18) = 0.19, p = 0.67	F(1,18.15) = 0.10, p = 0.75 T x S:
		T x S:	T x S:	T x S:	T x S:	T x S:	T x S:
		F(3,54) = 0.47, p = 0.71	F(3,54) = 0.02, p = 0.99	F(3,54) = 1.52, p = 0.22	F(3,54) = 0.21, p = 0.89	F(3,54) = 1.73, p = 0.17	F(3,54) = 2.74, p = 0.05	F(3,53.31) = 2.49, p = 0.07
	Non-enriched	0.59 ± 0.29	0.17 ± 0.17	0.22 ± 0.18	0.27 ± 0.26	0.25 ± 0.23	0.47 ± 0.24	0.55 ± 0.32
	Enriched	0.44 ± 0.25	0.20 ± 0.20	0.21 ± 0.23	0.29 ± 0.29	0.22 ± 0.21	0.48 ± 0.29	0.60 ± 0.35
	ST removal	0.49 ± 0.28	0.19 ± 0.16	0.19 ± 0.14	0.21 ± 0.22	0.25 ± 0.22	0.42 ± 0.26	0.44 ± 0.32
	LT removal	0.50 ± 0.28	0.18 ± 0.15	0.26 ± 0.19	0.24 ± 0.25	0.27 ± 0.25	0.51 ± 0.30	0.56 ± 0.34

For the proportion of total lying down that eyes were open, there was a significant interaction between treatment and sex in the Sept observation period (p = .04), with post-hoc tests showing the non-enriched females spent the least amount of time lying with their eyes open, although they did not differ from the non-enriched males. The main effect of treatment (p = .0003) showed the non-enriched mink spent the least amount of time lying with their eyes open (Table 3). There was a significant effect of sex at the Oct observation period (p = .03) with females showing less lying with their eyes open than males, but no effect of treatment (p = .35), and no interaction between treatment and sex (p = .38, Table 3). There were no significant effects of the treatment (p ≥.09), sex (p ≥ .11), or their interaction (p ≥ .11) at all other observation points (Table 3). Sleeping comprised 96.5% of the total observations of inactivity, with “resting, curled, eyes open,” comprising the next highest proportion at 2.84% of observations. There were no differences between treatment groups (all p ≥ .07), sex (p ≥ .42), or their interaction (p ≥ .05) in the proportion of time spent in the home cage (Table 3). Proportional values indicated mink spent numerically more time in the home cage in the observation periods of Sept, Jan and Feb (Table 3).

Across all treatment groups, the mean proportions of activity were observed to be numerically higher in the Jan and Feb observation periods, as were the mean proportions of stereotypy, but not of scrabbling (Table 3). Numerically greater proportions of time lying with eyes open were observed in the Sept and Oct Prior observation periods (Table 3). When mean stereotypy performance as a proportion of total activity was visually displayed across all observation periods by Family ID, there was clear visual variation in stereotypy between family groups (Figure 3).

Figure 3. The mean proportion (±SEM) of stereotypies (excluding scrabbling) as a proportion of total activity in mink from the different family groups across the entire trial.

The tail-scoring results showed that some mink from each treatment group chewed their tails at each of the four scoring days with the exception of the short-term removal group which showed no tail-chewing on the December scoring day (Figure 4). At the last tail-scoring date (February 2014) 35% of non-enriched mink, 10.5% of continuously enriched mink, 5% of short-term enrichment removal mink and 20% of long-term removal mink had some degree of tail damage (Figure 4). However, on the first tail-scoring date 20% of non-enriched mink, 5% of continuously enriched mink, 5% of short-term removal mink, and 30% of long-term removal mink showed some degree of tail damage. At the first tail-scoring date, the long-term removal mink had not yet had their enrichments taken away. At the final tail-scoring date, the non-enriched and long-term removal groups had more mink with more severe chewing scores (Figure 4), and the non-enriched mink showed the greatest mean length of tail damage (Table 4).

Figure 4. The percentages of mink from each of the four treatment groups (non-enriched, enriched, short-term removal, long-term removal) showing the 5 fur-chewing scores (depth of fur-chewing) on their tails as visually scored on four separate days (November, December, January, February). Nov immediately followed the first short-term removal observation period, Dec immediately followed the second short-term removal observation period and long-term removal, January and February correspond to the winter period after long-term removal. The scores are as follows: score 1: skin visible; score 2: very short but skin not visible; score 3: fur depth is ¼ chewed; score 4: fur depth is ½ chewed; score 5: no chewing.

Table 4. The mean (±SEM) estimated tail damage length (cm) for mink from four treatment groups (enriched, non-enriched, ST (short-term) removal, LT (long-term) removal) visually scored on four separate days (November, December, January, February). November immediately followed the first short-term removal observation period, December immediately followed the second short-term removal observation period and long-term removal, January, and February correspond to the winter period after long-term removal.

	Tail-scoring date
Group	Nov	Dec	Jan	Feb
Enriched	0.03 ± 0.03	0.45 ± 0.40	0.03 ± 0.03	0.79 ± 0.58
Non-enriched	0.65 ± 0.32	0.58 ± 0.38	2.0 ± 0.77	1.95 ± 0.75
ST removal	0.03 ± 0.03	0	0.48 ± 0.28	0.05 ± 0.05
LT removal	0.55 ± 0.25	0.53 ± 0.25	0.58 ± 0.34	0.6 ± 0.33

Discussion

This study aimed to determine the effects of simple manipulable cage enrichments and their removal, either short-term or longer-term, on the behavioral expression of welfare of fur-farmed mink in an experimental setting. Overall, there were few effects of these simple cage items or their removal on behavior across the treatment groups. Locomotor stereotypic and scrabbling behaviors were not reduced by the items nor affected by their removal. Observations conducted early in the trial (September) showed the non-enriched mink spent the least amount of time lying with their eyes open. Across the whole trial, this September observation period was generally when mink within all groups were recorded as showing relatively higher proportions of time lying with their eyes open. Patterns observed across time indicated all treatment groups showed increased activity in January and February, including increased locomotor stereotypies. Patterns of variation in locomotor stereotypies suggest differences across sibling groups in frequency of performance. No enrichment provision and its removal may have increased fur-chewing on the tail, but larger sample sizes would be needed to statistically confirm this given the low occurrence.

The two simple enrichments of a dumbbell and chewing chain showed few effects on mink behavior. Contrary to predictions, these manipulable cage items did not reduce stereotypic behavior. Lower stereotypies in enriched mink have been observed across a multitude of experimental studies with similar sample sizes as the current study, but the enrichment schemes have typically been more elaborate than what can be practically applied on-farm such as including running water and additional space along with manipulable items (Campbell et al., 2013; Dallaire et al., 2012; Díez-León et al., 2013, 2016; Meagher et al., 2013). Comparably simple enrichments of plastic balls and hanging chain or hanging hose on commercial farms in Canada across large numbers of individuals also did not result in reductions in stereotypic behaviors, although there were other positive effects of the cage items (Meagher et al., 2014). Thus, simple manipulable cage items may not be enough to counter the impacts of the commercial environment on abnormal behavior development. They may also need to be provided from an earlier age and with more variety to enhance novelty and use (Clark et al., 2023) which may increase positive enrichment impacts. It is unknown if there were any initial effects of these cage items on stereotypies in the first few weeks following provision as observations first occurred after several weeks of exposure. If stereotypies were initially reduced, this could indicate enrichment changeover to sustain novelty may be of benefit (Clark et al., 2023).

Stereotypies are prevalent abnormal behavior in the fur-farming industry indicative of underlying welfare issues. However, locomotor stereotypies themselves may not always be harmful to the individual, as they may be a coping mechanism (Mason & Latham, 2004) and have been shown to positively correlate with hippocampal neurogenesis (Malmkvist et al., 2012). Svendsen et al. (2007) showed genetic selection against stereotypies resulted in a higher proportion of fearful individuals, and high stereotyping females have been demonstrated to have higher fertility, possibly because the stereotypic activity reduced their body weight (Jeppesen et al., 2004). However, Díez-León et al. (2013) found that male mink that were more stereotypic were less successful at mating, which may have been due to the neurophysiological differences associated with the repetitive abnormal behavior. Stereotypy levels combined with other measures will likely provide a more robust indication of an individual’s welfare status than measuring stereotypies alone. In a large-scale commercial study on Canadian farms, other positive effects of enrichment were detected including increased juvenile play, reduced aggression and fear, more kits weaned, and on one farm, lower fecal cortisol metabolites (Meagher et al., 2014). If other behavioral and physiological variables had been assessed on the mink in the current study, such as a “glove test” for measuring fear and aggression or fecal cortisol metabolites, differences between treatment groups may have been detected.

There was an indication that a lack of enrichment increased the occurrence and severity of tail fur-chewing. This increased fur-chewing in the non-enriched group aligns with the reduced likelihood of fur-chewing in enriched mink on commercial farms (Meagher et al., 2014). More fur-chewing in the long-term removal group suggests enrichment removal does lead to frustration, but the low incidence of this behavior prevented any formal analyses. Additionally, at the November observation point (prior to enrichment removal), there were similar scores in fur-chewing between non-enriched and long-term removal mink suggesting inter-individual variation may play a strong role in the occurrence of the behavior.

Observations early in the trial showed there were some effects of the enrichment on inactivity subtypes within non-enriched mink; female non-enriched mink showed less time lying with their eyes open. Previous investigation into inactivity subtypes has shown variation between enriched/non-enriched mink, males and females, and location of inactivity (nest box or home cage). The variations were interpreted as inactivity subtypes indicative of welfare state. Lying awake across both locations may represent boredom and inactivity in the nest box may represent increased fear (Meagher et al., 2013). The findings of the current study conflict with this result, in that the non-enriched mink would have been expected to spend the most time lying awake if they were bored, not the least. It does, however, align with Polanco et al. (2021) who also (unexpectedly) found that lying awake increased in enriched mink. In the current study most inactivity subtypes occurred too infrequently for any statistical analyses to be performed, and the inactivity subtypes were not divided into the location in which they occurred (nest box or home cage). Most of the inactivity was observed as “sleeping” (curled up with eyes closed). This does align with the colder temperatures across several of the observation periods and where it may have been too cold for the mink to lie down stretched out with belly up or down.

North American commercial mink farming guidelines require manipulable enrichments and some pens (e.g., whelping pens) also require a shelf/tube/hammock within the cage (Fur Commission USA, 2019; National Farm Animal Care Council [NFACC], 2021). The combination of these enrichments may have greater positive impacts than manipulable cage items alone, or enrichment effects may be best detected during the breeding process (e.g., copulation or whelping success) rather than in the singly housed juvenile/young adult phase. Even if the manipulable cage items do not result in visible reductions in stereotypic behaviors, they may be creating positive experiences for the animals through the interaction with them. For example, these authors and authors of previous enrichment trials have anecdotally observed mink sleeping with their enrichments in their nest boxes suggesting they are valued items. This research contributes to the body of literature on possible enrichments for improving the welfare of fur-farmed mink, with evaluations conducted on simple, practical manipulable cage items relevant to commercial practice.

Acknowledgments

This research was funded by Fur Commission USA. Thank you to Angelo Napolitano and staff at the Michigan State University Experimental Fur Farm for technical assistance. Thank you to Bio-Serv® for donating enrichment samples. Appreciation is expressed toward the numerous undergraduate research interns within the Animal Behavior & Welfare Group at Michigan State University for assisting with data entry.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

All data supporting these results will be made available by request to the corresponding author.

Additional information

Funding

The work was supported by the Fur Commission USA.