There’s a new paper in Current Biology (click screenshot below, pdf here, and reference at bottom), which shows that, using standard methodology for measuring attachment to staff (“owners” in the vulgate), both kittens and adult cats are as attached to their staff as are human babies and d*gs. The paper is very short—two pages if you don’t read the supplementary material—and you can peruse it below. There’s also a good summary of the results in Ars Technica by Jennifer Oullette.
Initially, 70 kittens between 3 and 8 months old were tested for their degree of “security” using the Secure Base Effect test (SBT). As Jennifer notes, it’s done this way:
The felines spent two minutes in an unfamiliar room with their owner, who sat on an X in the middle of a circle marked on the floor and could only interact with their cat when the animal entered the circle. The owner then left the room, leaving the feline alone for two minutes, followed by a two-minute reunion, with the owner once again seated on the X within the circle. The researchers then examined the resulting video footage of how the cats behaved upon reunion with their owners, grouping them into three categories: secure attachment, insecure-ambivalent, and insecure-avoidant.
Securely attached cats actively greeted their owners upon reunion, yet were sufficiently calm and confident not to cling too much. “They’re able to use their owner as a secure base to explore out from,” said Vitale, with behavior evenly balanced between greeting their owner and exploring the unfamiliar space.
That’s in marked contrast to the behavior of cats deemed insecure-ambivalent. In that case, “We see persistent stress after the owner’s return,” said Vitale. “Rather than going back to exploring, they cling to their owner’s lap, sometimes engaging in a kind of spinning behavior where they almost can’t get comfortable in the owner’s lap.” An insecure-avoidant cat “shows very little visible response to their owner’s return,” said Vitale—not greeting their owner, ignoring them, or turning away.
Here’s a video showing how it’s done:
The results: Among the kittens, 64.3% were “securely attached” and 35.7% were insecurely attached, with the latter category comprising 84% ambivalent kittens, 12% avoidant kittens, and 4% “disorganized” kittens. The 64%/36% dichotomy is similar to that seen among human children (65%/35%) and d*gs (58%/42%: d*gs are MORE insecure, as expected)—all using the SBT protocol.
To see if the attachment could be increased by training, some of the kittens were given a six-week “training and socialization intervention with their caretaker”. These lessons had little effect: there were no differences between experimental and control (un-tutored kittens) after the training or even two months later.
Further, the researchers did the same test on adult cats, and the distribution of secure and insecure cats was pretty much the same as the kittens. Here’s a graph showing the baseline kittens, “socialized-trained” kittens, and adult cats. No standard errors are given, but the distributions look pretty much the same. And there’s also a photo showing the different attachment styles:
(Why is the “insecure-ambivalent” attachment sitting in the staff’s lap? Because it’s clingy, and seeking reassurance.)
Finally, the authors measured “separation distress” in cats, as quantified by “meow vocalizations” when the owner was gone. As they report ”
Attachment theory predicts that both secure and insecure-ambivalent individuals should show signs of distress including vocalizations during this phase with little difference between secure and insecure-ambivalent individuals . Insecure-avoidant and disorganized individuals should show less separation distress . As predicted, cats categorized as secure or insecure-ambivalent showed equivalent levels of separation distress in the alone phase as measured by vocalization frequency (U(32) = 68.5, Z = –0.843, p = 0.40; see Video S1 Timestamp 8:08), and both groups showed significantly more attachment distress than insecure-avoidant and disorganized cats (H(2) = 7.39, p = 0.025).
Oh, those disorganized cats! I’m not sure why, however, secure versus insecure-ambivalent cats shouldn’t differ in their levels of distress. The authors say that this is predicted by “attachment theory”, but that’s above my pay grade.
In the end, one concludes, at least from this protocol, that cats are just as attached to their staff as are babies and d*gs. The authors suggest that this is because cats and d*gs are both facultatively social “in free-roaming settings”, and fluctuate between being solitary and running in groups. Cat, like dogs, are “social generalists,” and the auhtors suggest that this flexibility has adapted cats to living in “anthropogenic environments”, a fancy-schmancy way of saying “living around humans.” Why is flexibility favored? I suppose because it’s good to get attached to some owners but also to retain some solitary traits, perhaps for purposes of hunting, or to deal with owners who aren’t so nice.
h/t: Bill, Michael, Dom et al. (many readers sent me this link; thanks to all)
Vitale, K. R., A. C. Behnke, and M. A. R. Udell. 2019. Attachment bonds between domestic cats and humans. Current Biology 29:R864-R865.