English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Jonathan Birch suggests that we should take one well-conducted study that produces results consistent with the idea of pain as being sufficient to invoke the animal sentience precautionary principle. Here, I consider how to balance negative and positive results from such studies using examples from my own work. I also consider which criteria of pain might provide strong inference about pain and which may prove to be weaker. Robert W. Elwood, Emeritus and Honorary Professor of Animal Behaviour, Queen’s University, Belfast, works on informationgathering, motivation, aggression and pain. scholar.google.co.uk/citations?user=JoEJgSUAAAAJ&hl=en Birch (2017a) considers the “Animal Sentience Precautionary Principle” and argues that one wellconducted study that is consistent with the idea of a taxon being able to feel pain should be sufficient to grant protection for the order within which that species is placed. He refers to work by me on decapods to illustrate his thoughts. I focus here on what to make of apparently negative findings and apparently positive findings. Some of the points I make here are touched upon in an excellent response to commentaries by Birch (2017b). In some cases, similar conclusions are reached; however, I go into more depth on specific issues, and I wrote before Birch (2017b) was made available. Are negative findings a problem? Mallatt (2017) and Woodruff (2017) are concerned by studies that fail to support the idea of pain in crustaceans and appear to believe that a study with negative findings should cancel one (or more) that has positive results. Lack of support is typically due to statistically non-significant results, whereas support is typically based on significant results. These outcomes, however, do not provide equal support for inference. A significant difference between experimental groups tells us that that under the conditions of the experiment, the animal is affected by the different conditions imposed by the experimenter. It provides strong inference. By contrast, a nonsignificant result does not allow the inference that the species in question cannot be affected by those conditions — just that the sample population failed to respond at that time. Thus, as a matter of logic, one negative result should not cancel a positive result. Absence of evidence is not evidence of absence (Sherwin 2001). Animal Sentience 2017.085: Elwood on Birch on Precautionary Principle 2 Woodruff notes that in my own work, two studies failed to support the expectations of pain. Magee and Elwood (2013) reported swift discrimination by shore crabs between a safe dark shelter and one that resulted in electric shocks, whereas another study (Magee and Elwood 2016), described by Woodruff as being ‘slightly different’, did not result in discrimination. Whilst the apparatus was broadly similar in the two experiments, the slight difference resulted in a very different paradigm. In the first, the animal was repeatedly placed in an enclosure containing the two shelters and had access to both during each trial. By contrast, in the second experiment, an opaque partition was placed between the shelters during each training trial so that the animal could only access one per training trial. In the first experiment, we had simultaneous presentation of the stimuli, whereas in the second, we had sequential presentation (eventually followed by a test when the crab could choose). Sequential presentation of stimuli in a discrimination task has been shown to be a more demanding task cognitively than simultaneous presentation in a variety of contexts (Dyer and Neumeyer 2005; Dougherty and Shuker 2015). The negative finding with respect to discrimination thus tells us more about the cognitive abilities of the animals than their ability to experience pain. However, even though there was no discrimination learning in Magee and Elwood (2016), the animals used different strategies to reduce their exposure to shock. An increasing number exited the shock shelter with successive trials; a comparison of those that exited on the first and last trials showed they exited more quickly in the later trial, thus markedly reducing the number of shocks. That is, they largely solved the problem of shock avoidance by a different means. The second, apparently negative finding of concern to Woodruff was about the expectation that animals should show vertebrate-like responses to morphine in inducing analgesia (Bateson 1991). Two studies had shown that crabs given morphine become unresponsive to shock, because they did not move away (Maldonado and Miralto 1982; Lozada et al. 1988), and this was interpreted as crustaceans showing morphine-induced analgesia. However, crustaceans given morphine also fail to respond to visual stimuli (Tomsic and Maldonado 1990). This opens the possibility that the initial conclusion about a specific analgesia was wrong. It could be that morphine induces unresponsiveness to all stimuli (Barr and Elwood 2011). To test this, an experiment was devised in which a crab was offered the chance to enter a dark shelter in which they would receive a shock. If morphine acts as an analgesic, then those given morphine should be more likely to move into the shelter to escape the light because the effect of the shock would be reduced by the analgesia. However, crabs given morphine were less likely to enter the shelter than were control, saline-injected crabs. All crabs given morphine were unresponsive for about 30 minutes after injection and tended not to enter the shelter even if no shock was delivered. Thus, it was concluded that morphine did not have a specific analgesic effect. The key question is, should we expect all taxa to have the same physiological response to morphine? We know that different taxa have very different physiology. For example, nociceptors in some taxa respond to capsaicin whereas others do not (Elwood et al. 2016). Thus, it is not reasonable to expect the physiological responses of all invertebrates to be the same as those of mammals, and this applies to the way responses to noxious stimuli might be modulated. A similar argument can be used to counter the suggestion by some (e.g., Key 2016) that only animals with brain structures identical to those of humans can experience pain. We know that similar functions — for example, the ability to use visual information — can occur in different taxa with very different brain structures (Elwood 2011). The same principle may apply to emotional responses to stimuli. Woodruff’s Animal Sentience 2017.085: Elwood on Birch on Precautionary Principle 3 concerns on this point are hence unfounded: The results of the various studies on morphine in crustaceans are not fundamentally different because they all indicate a general lack of responsiveness. The results did not differ, although they were interpreted differently. Which criteria should be regarded as important? The second point I wish to consider is Birch’s suggestion that only one study consistent with the idea of pain should be required to trigger protection under the precautionary principle. If it is to be one study, then it should investigate a criterion of pain that is very convincing (Sneddon et al. 2014). Birch picks motivational trade-offs as a good criterion of pain. The original aim of examining trade-offs was to establish whether responses were reflexes or more than reflexes. If they are reflexes, then responses should be the same regardless of other motivational conditions. However, the finding that hermit crabs are less likely to evacuate from a shell in which they receive a shock when the shell is of good quality (Elwood & Appel 2009), or when there are clues to the presence of predators (Magee & Elwood 2016), demonstrates that the responses are influenced by other motivational requirements and are hence not purely reflexive in nature. These experiments indicate that the noxious stimulus of shock produces a motivation to avoid the shock that is traded-off against other requirements, such as the requirement of a good quality shell or the need to avoid predators. Importantly, these and other experiments also indicate that the shock produces a motivation that appears to be strong because highly valuable resources may be abandoned to escape the shock. For example, hermit crabs that leave their shells are at risk of predation (Appel & Elwood 2009a), as are shore crabs that abandon dark shelters and move into a light environment (Magee & Elwood 2013). Trade-offs, together with giving up valued resources, are consistent with expectations of pain. This approach enables us to judge the price the animal will pay to avoid the noxious stimulus. Self-administration of analgesics provides a compelling case for pain experience (Colpaert et al. 1980), but there are problems with this approach in crustaceans, not least because it is not clear what might be an effective analgesic should the animals experience pain. I am not aware of any attempts with crustaceans, but one attempt with honey bees given the opportunity to ingest morphine did not find that their preference for morphine increased when they were injured (Groening et al. 2017). This might be because a) bees do not experience pain; b) they do feel pain but ingested morphine does not act as an analgesic; or c) they do feel pain, and morphine acts as an analgesic but the association was not formed because the bees could sample morphine and non-morphine solutions within the time that it took for morphine to act. It is reasonable to assume that the ability to experience pain evolved because it enhances long-term fitness. One way pain might be adaptive is that it induces or facilitates avoidance learning of the noxious, potentially damaging stimulus. However, mere demonstration of avoidance learning is weak evidence of pain. Decerebrate shore crabs, for example, appear to learn to keep a leg out of shallow water to avoid shock but only after receiving over 1,500 shocks in t

Assessing negative and positive evidence for animal pain