Aidan3

so the dictionary is wrong? Nope. Neither are "wrong", that isn't where the problem lies.

Anyone who'd say that many sighthounds aren't willing to take risks has never seen them after prey. And that includes a lot of pet owners in the burbs. To be honest I've not noticed that they are not particularly interested in approaching novel objects though? Definitely willing to take risks in pursuit of prey, but they aren't usually looking for trouble unless it's moving away from them. That's the tricky part in all this, we have to accept that the boxes don't fit everyone. Believe me, psych classes are full of students picking the fault with every little detail of every definition or theory we are given - and it is encouraged and expected. We also have to find the stuff we agree with, too, of course...

And here we have the fundamental problem with communicating these concepts, Rebanne. Does the dictionary definition look the same as the one I posted? Not really.

From Svartberg & Forkman (2002): "The shyness–boldness axis, a fundamental dimension in humans that can be defined as an individual’s general tendency to approach novel objects and willingness to take risks (Kagan et al., 1988; Wilson et al., 1994)" I have no idea what the survey questions were, but earlier corvus mentioned that they were also validated (meaning, someone has tested them to see if they convey the same information reliably).

So what's the point of attempting to describe personality dimensions across an entire group of breeds? I'm not being a smart arse - I genuinely don't get the point. What's the point of attempting to draw conclusions about the personality of an Irish Wolfhound based on surveys completed by Greyhound owners??? I'm not sure that corvus will do that. Her computer told her it found something, she's now trying to find out why it found something. Some of the discussion here told her exactly where she needed to look, I suspect.

Agreed, it's the same good point that Greytmate made and I echoed earlier. Communication is important. I'm lucky, I can see both sides, which is why I'm going to the trouble to explain (as best I can) the difference between the perception and the reality.

I'm seeing aggravation but I'm not seeing paranoia. I'm seeing both. If people are worried about what happens if this ends up in the wrong hands, then that is paranoia. It's certainly justifiable to worry about that sort of thing (it can happen), but this time? Seems implausible to me and I've outlined my reasoning.

The point is it's not a new axis. It's a well defined one that has been validated over and over from many different angles. A construct should fit with stuff it would be expected to fit with, and not fit with stuff you wouldn't expect it to fit with, and predict things you'd expect it to predict, and explain stuff you'd expect it to explain. Construct validity is a very well developed field of science in it's own right. When we're talking about personality dimensions, we use constructs that describe them, "boldness/shyness axis" for e.g A breed standard is what dog breeders of just one breed are supposed to be breeding towards. It does not describe personality dimensions across all breeds within a species. A researcher could pull out an adjective from a breed standard and work with it, but no doubt the word "aloof" (for e.g) already fits within some existing construct which has been validated.

Those tired old boxes are very useful. By making those boxes you stop a lot of hand-waving and speculation; you have a clearly defined tool. I think most researchers recognise that they are also limiting, contrived, and that no individual fits neatly into a [metaphorical] box. It's a bit like "positive reinforcement". We only call it that because we need a label. Giving a dog some food, and tossing a ball are completely different things, but we apply the same label (if it fits the definition). We recognise that the difference between tossing a treat and tossing a ball matters, but it still gets put in the same box. We can research positive reinforcement fairly confidently and draw conclusions about it and everyone knows what we are talking about.

So as that axis exists, we better make damn sure we find a way to jam these pesky hounds into that box. Otherwise, if they don't fit into the pre determined axis?? Total chaos may well reign... On the contrary. If corvus thinks she could show a way that those pesky hounds DON'T fit onto that axis she would be rubbing her hands together with glee! That would get published. More likely is the possibility that her research methodology didn't take into account certain factors, but it would be foolish to make that assumption.

Kinsella, you have nothing to worry about, this is not a paper, it's a survey. You put all the data into the computer and the computer says "hey, look at this!". Then you have to figure out why there is a number with too many zeros to ignore. That often doesn't lead anywhere. I think the topic title is regrettable, and corvus could have communicated these concepts better, but the level of paranoia is way out of proportion to the threat to anyone's breed group.

That's the point, though, the boldness/shyness axis would have been used (I say 'would have' because I'm not privy to these details so don't take my word for it) because it's already out there in the scientific literature. Researchers can and do challenge the existing literature every time they undertake a research project, to either support or reject what has gone before in some way. Maybe by finding that sighthounds are not "bold", but that they also aren't "shy", you challenge that construct? Corvus wishes! More likely there is a problem with the way the data was collected (e.g because they are racing greyhounds).

That's probably true, but corvus will need to draw on validated, previous research which uses terms such as the "boldness/shyness axis". Construct validation is painstaking work.

I wish I shared your confidence Aidan. Clearly the OP doesn't think its implausible. To quote from page 1: "The most timid group is toy breeds. That kinda makes sense to me because they are very small and aren't really expected to handle much pressure. The boldest group is guardian breeds, which makes sense to me because they are very large and you would think guardian breeds would do better if they were more bold than timid. But sighthounds... Not sure about that one" The other thing to understand is that this is a doctoral dissertation. Sometimes these things can attract a bit of attention, especially if the author is good at self-promotion and full of enthusiasm, but it would take a brave academic to begin their career by trumpeting the results of self-report on-line survey data. Especially when corvus has a much more interesting component to her research into dog optimism that relies on experimental data.

No-one says anything about self-report surveys so I think the concerns that people have are blown way out of proportion. Oprah is not going to be telling us all that sighthounds are "timid killing machines on a leash" next week as she introduces her nobel prize winning guest, corvus You take a survey when you want a lot of data on the cheap. Then you want to drill down and figure out which factors appear to be significant. Then, if you are inclined, you do something more objective to determine the truth of the significant bits. To do this you start with the idea that what you saw was a chance event, someone farted in the wrong direction and stank up your data. You can't infer anything from self-report survey data. Especially on-line data. Especially on-line data taken from members of special interest fora. It's essentially exploratory data. The thread topic is regrettable, but who is going to take that seriously? It becomes evident from page 1 that this is a conception of an idea extrapolated from the data that requires extra input, because on face-value it seems implausible.

If the construct ("timidity") isn't valid, nothing will be published and corvus knows this. It's one of the many, many pieces she will have to put together before she gets anywhere which is, I imagine, why she asked here. In between all the speculation she got some pretty good answers as to why owners of sighthounds are reporting that their dogs are not bold. and what's the definition of bold? This is one of the problems with self-report surveys, you more or less rely on the individuals interpretation of the survey questions (or worse, what they think they should say). That's one of the reasons why when you fill in one of these surveys you'll feel like you're answering the same question, worded differently, over and over. The good news is that anyone who will do anything with the data knows this and no-one will infer anything from it other than "that's interesting, we could look more closely at that using methods designed to test it more objectively". How the responses of owners of mostly ex-racing greyhounds can be expanded to cover a group of dogs bred on different continents for different purposes and with different whelping and early environmental experience might be one threshold question to address. The fact that all sighthounds share some physical similarities doesn't mean they were bred from the same genetic material or that they performed identical functions. Comparing the early experiences of these animals would be different and comparing dogs that still largely race (and not much else) to breeds that have been largely bred as companions for some time are also variables that the conclusion doesn't address. A timid greyhound with healthy prey drive will still race well enough. I think its the generalisation that's getting up people's noses, not the inference. Some sighthounds ARE timid. Fact is they shouldn't be and many of us have ones that arent. Starting with a study group skewed in favour of a particular breed selectively bred for one characteristic only and raised a particular way is going to provide atypical answers for the group of breeds as a whole. If the generalisation is getting up this population's nose, you can only imagine how badly it would be savaged in academia The sort of statistical analysis that corvus is using picks this sort of thing up pretty easily so I'm not going to jump the gun in assuming that the other sighthounds in the study were any different to the greyhounds. How many other breeds were represented? I don't know. It's something for corvus to address in her paper's discussion section.

Good find, thanks for sharing.

If the construct ("timidity") isn't valid, nothing will be published and corvus knows this. It's one of the many, many pieces she will have to put together before she gets anywhere which is, I imagine, why she asked here. In between all the speculation she got some pretty good answers as to why owners of sighthounds are reporting that their dogs are not bold. and what's the definition of bold? This is one of the problems with self-report surveys, you more or less rely on the individuals interpretation of the survey questions (or worse, what they think they should say). That's one of the reasons why when you fill in one of these surveys you'll feel like you're answering the same question, worded differently, over and over. The good news is that anyone who will do anything with the data knows this and no-one will infer anything from it other than "that's interesting, we could look more closely at that using methods designed to test it more objectively".

If the construct ("timidity") isn't valid, nothing will be published and corvus knows this. It's one of the many, many pieces she will have to put together before she gets anywhere which is, I imagine, why she asked here. In between all the speculation she got some pretty good answers as to why owners of sighthounds are reporting that their dogs are not bold.

Academic speeak for " we make it up as we go along " We generally prefer to start with a preconceived idea then try to shape our data to fit that. Making it up as we go along allows the data to have it's own say (like here), and that doesn't get the really big grants.

It sounds very unusual for a reputable organisation.

My immediate thought is that, because the magnitude of the reinforcer changed unexpectedly, they might be exploring to see what it is that caused the change. If you press a button 200 times and get a single M&M on whatever schedule there is in place, you have an expectation of a single M&M. Then all of a sudden you start getting 10 M&Ms, you're going to wonder what you did. If you want to get a visual representation of what happens to any associated respondents, this java applet allows you to run repeated trials while playing around with the strength of the unconditioned stimulus. The output is determined by the Rescorla-Wagner model. You can also investigate blocking and shadowing effects using different stimuli, which might be a consideration if you 'step back' and view your apparatus from the dog's perspective.

Or live 500km from anyone competent...

Good work, Sheena

I've never noticed this happen, but I'm imagining that if a ball or kong on a rope slips out of the mouth the teeth could knock together. It would depend a lot on the dentition I guess.