Canine Ergonomics 1 & 2 (Patreon Book Club Notes)

As part of our monthly Patreon Benefits, we have a book club. For the next few months (summer to fall 2022), we’re reading Canine Ergonomics by William Helton. These rough notes are for chapters 1 and 2 of the book and include heavy quotes of notable passages from the book.

Join our book club on Patreon now to hop on the next call!

Chapter 1: Canine Ergonomics (1-16)

• Some claims here w/o citations – eg, we aren’t descended FROM chimps (and he does not even mention bonobos), but we share a common ancestor.
• “People have excellent sensory capabilities in daylight because of stereoscopic color vision, but relatively low levels of night vision, olfactory ability, and hearing put people at a distinct disadvantage in darkness. Wolves in the dark or during dawn and dusk benefit from superior night vision, exceptional olfactory abilities, and better hearing.”
• Fear of anthropomorphism = anthropodenial
• “Anthropomorphism is indispensable with working dogs because they are, by definition, anthropomorphic.” WOAH.
• “Working dogs often act as surrogate humans” by doing jobs humans cannot (sniff bombs) or replacing humans for tedious jobs that would be unaffordable for a person (guide dog)
• Is this true?? “Dogs, like humans, are products of uncontrolled evolution—they were not built with a purpose in mind.”
• “Both dogs and humans descended from social predators with hierarchical societies.” Frans de Waal seems to dispute human end given our equal closeness to bonobos as chimps. Not sure about dog side either (wolves are not very hierarchical).
• “Working dogs are unique because of the ease with which they work in complex human settings and integrate into human society.”
  • (1) working dogs are nonhuman models of human workers; more generally, working dogs and working people share much in common; (2) the study of working dogs can improve the science of ergonomics as the study of nonhuman models led to advances in other areas of behavioral science; and (3) an ergonomic orientation, recognizing dogs as real workers, can benefit working dogs. 
• Dogs can help as a model for debates in talent (genetics) versus learning because you cannot force humans to endure long studies that would examine this in people – and those that didn’t opt out may stay in due to genetic talents that make them succeed more.
  • Note: It seems that selective breeding, ENS, temperament tests, and all the rest that goes into itty-bitty puppies would confound this claim
• Novice stage: acquiring declarative knowledge, explicit understanding of how to perform a task. At this stage, performance is mentally taxing. 
• Language =/= learning
• Slabbert and Rasa try to demonstrate observational learning in narcotics puppies, but the summary doesn’t mention the extreme stress of removing puppies from maternal care at six weeks?!
• “Animal behaviorists studying canine workers are aware of potential limitations in the dogs’ capacity to sustain effort over time; they, however, tend to interpret decrements in performance efficiency as motivational problems (e.g., boredom), not fundamental limitations of attention resources (Garner et al., 2000)”
• Conscious awareness of a rule (and ability to articulate it) are not the same as expertise

Chapter 2: Skill and Expertise – Cognitive Approach (17-36)

• Use of the expertise term in relation to dogs is contentious because of the connotation and implication nonhuman expertise has for animal consciousness, the relative contributions of nature and nurture to development, and the allocation of research funding to technology 
• As early as 945 AD, the Welsh laws of Hywel Dda formally declared the difference between expert dogs and untrained dogs by imposing different legal penalties for killing them 
• Dogs can present in court: expertise?
• Social Construct: Other social markers of expertise such as honors, medals, awards, and titles, for example, are useful
• What can experts objectively do and do better than their more ordinary colleagues? 
• Performance: Most dog skills, moreover, meet Ericsson’s (1996) three criteria of study by scientific methods: they are reliable, reproducible, and clearly describable 
• Extractable Knowledge: This reliance on verbal reports by researchers advocating this definition of expertise makes such investigation of nonhumans difficult. This perspective intermingles language and communication capacity with expertise and this conflation is unnecessary. 
• “The ability to locate hidden items and the communication of such information implies that dogs are capable of communicating mental representations to others. While this method for inferring mental representation may not be exactly the kind of knowledge representation that cognitive engineers seek, it indicates that dogs can use abstracted knowledge.”
• “The reliance on verbal reports by advocates of the knowledge definition of expertise may imply a close tie between ability to verbalize expertise and actual expertise, but this conflation is a fundamental categorical mistake. It confuses being an expert with being able to talk about ways in which one is an expert.”
• Is awareness of a heuristic necessary to use it? Obviously no, but is it the same thing then??
• “Outcome of Process: Deliberate Practice: the cognitive features of deliberate practice are heavily laden with assumptions about conscious states, and this phenomenology is problematic for assessing dogs The behavioral features of deliberate practice include motivation of a trainee to improve, assignment of well-defined tasks, delivery of feedback to the trainee, and ample opportunity for repetition” 
• “Denying a dog an opportunity to work is punishment. The other behavioral features of deliberate practice are simply descriptive facts of dog training.”
• Level of expertise may reflect the end result of this process, when performance reaches near an asymptote of ability. 
• “Outcome of Process: Lifelong Learning. expertise can be defined as the outcome of sustained learning occurring over at least 10% of the species-typical lifespan. This definition is very similar to the deliberate practice definition, but attempts to diffuse some of the angst scientists exhibit about “deliberate” actions of animals. “
  • Greyhound example is a bit far-fetched to me; it could definitely be practice and training but also isn’t physique/physical development partially to blame for similar peaks in performance in relation to lifespan?
• If expertise comes from deliberate practice, does expertise in an animal equal consciousness?
• Phenomenal consciousness is the subjective experience of mental life or the feeling of being conscious. 
• “Access consciousness, conversely, concerns the availability of mental representations necessary for the cogent control of action and can be more formally described in information processing terminology”
• “In an initial cognitive stage, the human trainee pays close attention to cues and feedback. Performance during this initial stage is irregular and requires active coordination of separate skill components. Skill production demands attention of the trainee. Over time and after practice, a trainee moves to an associative second stage consisting of the organization of separate skill components into larger units or chunks. This stage is characterized by a noticeable increase in skill fluidity and a decrease in the need for active attention. With more practice, the trainee shifts to a third autonomous stage in which the skill becomes essentially automatic, independent from cognitive control and attention. Skill automaticity frees the trainee’s attention resources for performing other concomitant tasks and monitoring performance of the central task.”
• “Ethological research on wild cheetahs provides some evidence supportive of a stage model of animal skill development (Caro, 1994). Cheetah mothers provide their offspring with opportunities to hunt in relatively controlled environments. They capture live prey and bring the prey back to their cubs. They then release the prey, presumably, for the cubs to chase. If the cubs are unable to capture the prey, the mother recaptures it, returns to the cubs, and again releases the prey. The cubs develop the physical skills of predation without additional cognitive burdens such as detection and decision making. Moreover, young cheetahs appear to be less capable of determining when a chase is futile, on average abandoning a chase after 18 m. An experienced adult cheetah will abandon a chase after only 2 m (Caro, 1994) and may devote more attention to detecting prey signals, because its physical skills have become automatic and demand less attention.”
• Agility is a similar model: can dogs both run the obstacles and perceive the handler’s cues, especially verbal cues and hand cues aside from mistaken flailing while running? (Same goes for handlers: can they remember the course and cue the dog at the same time while also avoiding inadvertent hand gestures, etc)
• Animal researchers could make greater use of this dual-task data derived from human expertise research (Beilock et al., 2002; Leavitt, 1979; Smith and Chamberlin, 1992) to investigate skill automaticity. 
• An interesting question remains: if skills in dogs become automatic, then what were they before they became automatic? 
• From a pragmatic perspective, individuals working with dogs may need to regard the dogs as conscious agents and let the philosophers deal with the nitpicky details of what consciousness means. 
• “Helton did not find any breed differences in agility precision or in the amount of training necessary to achieve mastery.” This seems suspect given BC domination in the sport?
• By selecting for physical shapes, one can influence perceptual systems, and these will, in turn, influence what appears to be cognition.