Finishing up the semester’s grading and preparing to leave for a little bit of fieldwork, but here is what I am trying to read today:
Let’s Abandon Significance Tests – Jim Wood (The Mermaid’s Tale)
But now suppose we’ve learned our lesson: and so, chastened, we abandon our arbitrary threshold a value and look instead at the exact p value associated with our predictor variable, as many writers have advocated. And let’s supposed that it is impressively low, say p = 0.00073. We conclude, correctly, that if the null hypothesis were true (which we never really believed in the first place) then the data we actually obtained in our sample would have been pretty unlikely. So, following standard practice, we conclude that the probability that the null hypothesis is true is only 0.00073. Right? Wrong. We have confused the probability of the data if you are given the hypothesis, P(Data|H0), which is p, with its inverse probability P(H0|Data), the probability of the hypothesis if you are given the data, which is something else entirely. Ironically, we can compute the inverse probability from the original probability – but only if we adopt a Bayesian approach that allows for “subjective” probabilities. That approach says that you begin the study of some prior belief (expressed as a probability) in a given hypothesis, and adjust that in light of your new data.
Alas, the whole NHST framework is by definition frequentist (that means it interprets your results as if you could do the same study countless times and your data are but one such realization) and does not permit the inversion of probabilities, which can only be done by invoking that pesky Bayes’s theorem that drives frequentists nuts. In the frequentist worldview, the null hypothesis is either true or false, period; it cannot have an intermediate probability assigned to it. Which, of course, means that 1 – P(H0|Data), the probability that the alternative hypothesis is correct, is also undefined. In other words, if we do NHST, we have no warrant to conclude that either the null or the alternative hypothesis is true or false, or even likely or unlikely for that matter. To quote Jacob Cohen (1994), “The earth is round (p < 0.05).” Think about it.
Hypothesis testing in paleoanthropology is essentially always plagued, whatever approach you take, by issues of sample size. This means that what we conclude tells us an awful lot about the approach we have taken and the samples we have used, but less clearly about what we are actually trying to find out. The calculation of Bayesian prior probabilities, however, is often plagued by the same problem as frequentist statistics given the constraints inherent to fossil studies. These are good and important conversations to have, but more than anything, I think they highlight the need to critically examine (and re-examine) the analytical approaches underlying fossil studies.
Biological Anthropology Resources…
The Lawn Chair Anthropologist (aka, Zach Cofran), passes along two links to online bioanth resources:
Online skeletal and dental datasets
Human Origins Database
And just in time for my own departure for the field, Ken Weiss starts a conversation on what we can do about field safety, given the issues raised by Clancy, et al..
Sexual Harassment in the Field (of Anthropology) – Ken Weiss (The Mermaid’s Tale)
So, if people really care about this subject, as it clearly seems they should, then what is needed is to try to find some way to formulate policies and procedures that might actually work. Discussions about how awful the problem is are fine, but realistically implementable ways to constrain action in unusual, hard-to-monitor settings is what needs attention. If the ongoing discussion since the Anthropology meetings has brought that attention to this issue, great. It clearly needs to continue.
Research Articles:
Jaw growth in the absence of teeth: the developmental morphology of edentulous mandibles using the p63 mouse mutant (Paradis, et al., Evolution & Development)
Abstract: Mammalian tooth and jaw development must be coordinated well enough that these systems continue to function together properly throughout growth, thus optimizing an animal’s survival and fitness, as well as a species’ success. The persistent question is how teeth and jaws remain developmentally and functionally viable despite sometimes monumental evolutionary changes to tooth and jaw shape and size. Here we used the p63 mouse mutant to test the effect of tooth development — or the lack thereof — on normal mandible developmental morphology. Using 3D geometric morphometrics, we compared for the first time mandible shape among mice with normal tooth and jaw development against p63 double knock-out mice, with failed tooth development but apparently normal jaw development. Mandible shape differed statistically between toothless (p63−/−) and toothed (p63+/−, p63+/+) mice as early as embryonic day (E) 18. As expected, most of the shape difference in the p63−/− mandibles was due to underdeveloped alveolar bone related to arrested odontogenesis in these E18-aged mice. Mandible shape did not differ statistically between p63+/− and p63+/+ adult mice, which showed normal tooth development. Our results support the idea of a gene regulatory network that is exclusive to the mandible and independent of the dentition. This study also underscores the biomechanical impact of the teeth on the developing alveolar bone. Importantly, this work shows quantitatively that the p63 mutant is an apt model with which to study mandible morphogenesis in isolation of odontogenesis to clarify developmental relationships between the teeth and jaws.
A comparison of antemortem tooth loss in human hunter-gatherers and non-human catarrhines: Implications for the identification of behavioral evolution in the human fossil record (Gilmore, American Journal of Physical Anthropology)
Abstract: Middle and Late Pleistocene fossil hominin specimens with severe antemortem tooth loss are often regarded as evidence for the precocious evolution of human-like behaviors, such as conspecific care or cooking, in ancient hominin species. The goal of this project was to ask whether the theoretical association between antemortem tooth loss and uniquely human behaviors is supported empirically in a large skeletal sample of human hunter-gatherers, chimpanzees, orangutans, and baboons. Binomial regression modeling in a Bayesian framework allows for the investigation of the effects of tooth class, genus, age, and sex on the likelihood of tooth loss. The results strongly suggest that modern humans experience more antemortem tooth loss than non-human primates and identify age in years as an important predictor. Once age is accounted for, the difference between the humans and the closest non-human genus (chimpanzees) is less pronounced; humans are still more likely on average to experience antemortem tooth loss though 95% uncertainty envelopes around the average prediction for each genus show some overlap. These analyses support theoretical links between antemortem tooth loss and modern human characteristics; humans’ significantly longer life history and a positive correlation between age and antemortem tooth loss explain, in part, the reason why humans are more likely to experience tooth loss than non-human primates, but the results do not exclude behavioral differences as a contributing factor.
Earliest Archaeological Evidence of Persistent Hominin Carnivory (Ferraro, et al., PLOSone)
Abstract: The emergence of lithic technology by ~2.6 million years ago (Ma) is often interpreted as a correlate of increasingly recurrent hominin acquisition and consumption of animal remains. Associated faunal evidence, however, is poorly preserved prior to ~1.8 Ma, limiting our understanding of early archaeological (Oldowan) hominin carnivory. Here, we detail three large well-preserved zooarchaeological assemblages from Kanjera South, Kenya. The assemblages date to ~2.0 Ma, pre-dating all previously published archaeofaunas of appreciable size. At Kanjera, there is clear evidence that Oldowan hominins acquired and processed numerous, relatively complete, small ungulate carcasses. Moreover, they had at least occasional access to the fleshed remains of larger, wildebeest-sized animals. The overall record of hominin activities is consistent through the stratified sequence – spanning hundreds to thousands of years – and provides the earliest archaeological evidence of sustained hominin involvement with fleshed animal remains (i.e., persistent carnivory), a foraging adaptation central to many models of hominin evolution.
