Is experimental psychology science?

But are you interested in predicting the specific behaviors of specific people at specific instances in time? I suppose it could be useful, say, in betting on sports events, but is it crucial theoretically? Who cares if we can predict that I respond to signal X at time Y in precisely 242ms?Just because there is variability doesn't mean we don't "understand" a behavior, or can't characterize it.Consider Fitt's Law.http://ei.cs.vt.edu/~cs5724/g1/glance.htmlHere's a well supported, mathematical description of human action. It's not 100% deterministic, but it's quite real. There's variability in real life measures of the path of a ball, too. The idealized physical equations don't translate into perfect real world precision. I dare you to go out into a field, apply force to a ball, and make it behave precisely the same each and every time you apply force. Why is this so important? Digit spans are "7 plus or minus 2", the so-called "Magic Number" of cognition. Any theory or model of cognition must account for the mean (7) AND the variability (plus or minus 2).Why is variability bad? It's data. It can be quantified. Variability is data, too. It tells us something about how the mind works. Any model that predicts a digit span of 7 AT ALL TIMES is wrong. It can't be an accurate model of cognition.This message has been edited by Zhimbo, 10-15-2005 05:02 PM

Certainly there is bad science within psychology, but repressed memory is the work of clinicians and social workers, NOT researchers. (In contrast to n-rays in physics, which was a mistake made by experimental physicists. Or cold fusion, also by experimental physicists. Why isn't Physics labeled "soft science", then?).Who has done the work that has debunked repressed memories? Research psychologists, Elizabeth Loftus being the most publicly vocal.Psychology has a bad rep primarily because everyone thinks of Dr. Phil, Freud, and Frasier. Or pop psychology like "Men are from Mars..." blah blah blah. Hardly anyone who poo-poos Psychology that I've come across seems to have even taken an intro course in the subject.nwr: by the way, I do realize your post was entirely reasonable. But "repressed memory" is something that Psychology - the science - has thoroughly debunked, and it irks me when people blame Psychology - the science - for the charlatanism of "counselors", many of whom have no training in psychology itself.ABE: For those who don't know about n-rays:Blondlot and N-rays - The Skeptic's Dictionary - Skepdic.comI find the following quite amusing in this context:

quote:

---

"But were those who verified Blondlot's N-ray experiments stupid or incompetent? Not necessarily, since the issue isn't one of intelligence or competence, but of the psychology of perception. "

---

So...to properly understand this failure within physics, you need psychology. Nice!This message has been edited by Zhimbo, 10-15-2005 06:07 PM

This kind of free-flowing hand waving is supposed to convince me that psychology isn't science?" but can they control their experiments? no."What does this mean? "Control their experiments"? Do you mean "use experimental control conditions"? What? Because for any reasonable, relevant meaning of this phrase, I assure you I can provide example of "control" in psychology."if psychologist want to do science, they'll explore the way the brain works, not how people feel."Of course, psychologists DO explore the way the brain works. And you know what? The brain works to produce feelings. Really. Honest. You can change how people feel by affecting the brain. So why are "feelings" not amenable to scientific inquiry?This message has been edited by Zhimbo, 10-15-2005 04:43 PM

There is that group of poo-poo-ers, too, although I think they're certainly a smaller group than people who just don't know what psychology is."In other words, people who are shocked that human beings aren't massless springs or frictionless pulleys."Yes...much of what I've seen here is people talking about how "variable" or "uncertain" human behavior is. Well, duh. If you then conclude that it isn't "science", then you haven't thought very much about what science really is. Your previous post was spot on.This message has been edited by Zhimbo, 10-15-2005 04:52 PM

Of course, the authors of that paper use the reaction time measurements to make inferences about aspects of cognition - mind. Is that paper science or pseudo-science?What's not physical: human motivations, feelings, ideas about morality, etc.So, any research that deals with motivations, feelings, or "ideas about morality" is, necessarily, pseudo-science? Is that your claim?

"psychology cannot restrict the variables. "Uh, yes it can? I must not understand what you mean. What are "the" variables?"and no, psychologists do not analyze how the brain works."Um, I'm a psychologist. And I most definitely analyze how the brain works."most psychologists refuse to 'reduce' psychological functioning to chemical and biological realities."OK, you're totally in some bizarro universe here. What are you basing this on? Why is there so much reference to the brain and its function in all these psychology papers I read (and write)? Seriously, not an issue of the pan-psychology journal Psychological Science goes by without without some article (if not several) relating psychological function to the brain. That's just an example.I've worked in three different Psychology labs. All of them were very concerned with relating psychology to biology and chemistry. And while not all labs do biologically-based work, this hostility you claim is present in "most" psychologists I seem to miss, despite working in the field.This message has been edited by Zhimbo, 10-15-2005 07:08 PMThis message has been edited by Zhimbo, 10-15-2005 07:11 PM

You didn't answer my question:So, any research that deals with motivations, feelings, or "ideas about morality" is, necessarily, pseudo-science? Is that your claim?Is that your claim? Yes, or no?

Well, there may well be isolated pockets of hostility towards biology, but as a *mainstream* attitude within psychology? No way. And it's not only "my world" - it's the mainstream psychology journals that I'm basing this on.Here are articles from the most-recent issues of mainstream, general-psychology (not biological psychology) journals. Articles are chosen only when they mention brain/biology/chemistry IN THE TITLE. I'm not reading all of the articles to see how many discuss biology in the text (from the topics I suspect many do).Latest issue of "Psychological Science":
"Frontal Brain Asymmetry and Reward Responsiveness. A Source-Localization Study"Latest issue of "Current Directions in Psychological Science":
"The Neural Bases of Placebo Effects in Pain"
"Components of Short-Term Memory and Their Relation to Language Processing. Evidence From Neuropsychology and Neuroimaging"
"Facilitation of Extinction of Conditioned Fear by D-Cycloserine. Implications for Psychotherapy"Latest issue of "Annual Review of Psychology"
"MODELS OF BRAIN FUNCTION IN NEUROIMAGING"
"BRAIN ORGANIZATION FOR MUSIC PROCESSING"
"VESTIBULAR, PROPRIOCEPTIVE, AND HAPTIC CONTRIBUTIONS TO SPATIAL ORIENTATION"
"THE NEUROSCIENCE OF MAMMALIAN ASSOCIATIVE LEARNING"
"BEHAVIORAL INHIBITION: Linking Biology and Behavior within a Developmental Framework"
"HUMAN DEVELOPMENT: Biological and Genetic Processes"
"THE PSYCHOLOGY AND NEUROBIOLOGY OF SUICIDAL BEHAVIOR"Latest issue of Journal of Experimental Psychology: General
"The Role of the Locus Coeruleus in Mediating the Attentional Blink: A Neurocomputational Theory."Latest issue of Psychological Review
"Integrating Incremental Learning and Episodic Memory Models of the Hippocampal Region. "Latest issue of Psychological Bulletin
"Evolution of Life-History Trade-Offs in Mate Attractiveness and Health: Comment on Weeden and Sabini (2005). "Of course, there are also tons of journals with titles like "Cognitive, Affective, and Behavioral Neuroscience", or "Brain and Behaviour", or "Cognitive Neuropsychology" that are certianly mainstream within Psychology, as well. Hardly an indication of a widespread reluctance to examine biological data in the course of psychological research.

OK, let me rephrase:Any "scientific" research that deals with motivations, feelings, or "ideas about morality" is, necessarily, NOT science but pseudo-science?That is, if a professional scientist publishes work in the journal, say, "Psychological Science", and talks about the scientific implications of the work...AND that work deals with motivations, feelings, or "ideas about morality"THEN that work is actually pseudoscience?

Assuming your answer to my previous post is "yes" - which I think it must be based on your posts so far - how about the following article which appeared in Nature, which mentions and deals with feelings (fear, judgments of trustworthiness, etc.):

quote:

---

The human amygdala in social judgment

​

RALPH ADOLPHS*, DANIEL TRANEL* & ANTONIO R. DAMASIO*”

​

Studies in animals have implicated the amygdala in emotional, and social, behaviours, especially those related to fear and aggression. Although lesion, and functional imaging, studies in humans have demonstrated the amygdala's participation in recognizing emotional facial expressions, its role in human social behaviour has remained unclear. We report here our investigation into the hypothesis that the human amygdala is required for accurate social judgments of other individuals on the basis of their facial appearance. We asked three subjects with complete bilateral amygdala damage to judge faces of unfamiliar people with respect to two attributes important in real-life social encounters: approachability and trustworthiness. All three subjects judged unfamiliar individuals to be more approachable and more trustworthy than did control subjects. The impairment was most striking for faces to which normal subjects assign the most negative ratings: unapproachable and untrustworthy looking individuals. Additional investigations revealed that the impairment does not extend to judging verbal descriptions of people. The amygdala appears to be an important component of the neural systems that help retrieve socially relevant knowledge on the basis of facial appearance.

---

This is just the abstract. If you'd like more info, I can see if there's a free link to the article; if not, I can provide more information.However, based on what you seem to be telling me so far, you would have to qualify this as "pseudoscience" based on it dealing with "non-physical" stuff like "trustworthiness" and "fear" and "approachability".If, instead, you think you need more information before dismissing this as pseudoscience, then you need to retract your claims as you've stated them.This message has been edited by Zhimbo, 10-16-2005 02:47 PM

Actually, I thought the paper that I provided was quite accessible. In the abstract, only "amygdala" seems to be ultra-jargony. The layman may not be able to find the amygdala in the brain, but to understand the abstract one only need know that it's part of the brain. The rest seems pretty accessible to anyone with a passing familiarity with psychology. Well, in the animal studies mentioned in the abstract, "fear" in animal studies is operationally defined as "freezing" responses and the like. The human work is about "judgements of trustworthiness" and "approachability". This is operationalized by, well, having people make judgements of trustworthiness and approachability using a rating scale. Nothing too mysterious, and I suspect these are pretty closely related to what people mean in everyday use of the terms. Sophistry? Isn't this debate about whether experimental psychology scientific? Didn't robinrohan make a claim? Didn't I provide material that made him rethink his claim as too extreme?I'm more than willing to move on from this point, but robinrohan's extreme claim - that anything dealing with "non-physical" concepts like motivation, emotions, and "ideas about morality" can't be scientific - is a widespread belief among people who take the "nay" position of this debate topic. Are you familiar with Damasio's Somatic Marker Hypothesis? I'm pretty sure it DOES address a layman's ideas about "fear" and "judgment of approachability" and "judgement of trustworthiness". And your question seems to imply that operational definitions of emotion can't reflect "true" emotion, only behavior. True, behaviors are measured, but do you really think that people giveing ratings of approachability is unrelated to a person's judgement of approachability? Or that a cowering rat isn't feeling fear?And, the topic of this thread isn't whether the jargon of psychology maps well to layman's notions, anyway. That seems to be an entirely different topic.

It isn't interesting to you? Well...so? My point is that, despite all the hand wringing people were making about the "variability" of human behavior, one can still find useful mathematical characterizations of the behavior. Just because psychology tends to have larger error bars than physics doesn't make it intractable or pseudoscience.Fitt's Law is "real". It's testable. It's useful in engineering applications. So why is everyone all concerned about the "variability" of human behavior, and what it means about the scientific status of experimental psychology? What is "high" precision? Predictions in psychology tend to be less precise. So? That can make things more difficult, but is irrelevant to the topic of this thread. Larger error bars does not make something pseudoscience. Sure it's a "problem". Does that make experimental psychology a pseudoscience? Baddelley's model makes minimal empirical claims. Yes, it's scientific, although it's extremely limited. It's a gloss, but it's not an "utter" gloss. Well, depending on what you mean by "see folk psychology", this statement might be a tautology, or it could be patently false. There are concepts in "folk psychology" that I think are clearly amenable to scientific investigation (e.g., memory). Surely you aren't saying that 1) only computational models are predictive or 2) all computational models are predictive.Many non-computational models make predictions (the ToE?), and many computational models are merely descriptive, not predictive. Many computational models of cognition amount to little more than programming languages of such power that they can explain nearly any possible set of results. My feelings exactly. I do both behavioral work and neuroimaging work, and it truly irks me that neuroimaging work gets more "automatic" respect simply because it has pretty pictures of brains.This message has been edited by Zhimbo, 10-16-2005 08:58 PM

Regarding point 1:The choice of p-values within neuroimaging work is a thorny issue. One must be fair here: standard statistical tools weren't made with neuroimaging data in mind! Neuroimaging data has a MASSIVE multiple-corrections problem, and involves spatially and temporally correlated data, not independent data. The field is getting a lot better, with various techniques becoming more standardized (False Discovery Rate of p<.05 rather than the standard family-wise error rate, etc.).The paper in question does not fall prey to this. There's an initial p<.01 "filtering" of the data (anything voxel weaker than this is out of consideration), then small volumes are used with full correction for multiple comparisons.2. I disagree...First, for the hypothesis testing of the paper, only the pattern of dissociation is required to support the predictions (that in one contrast IFG is significant, but ACC isn't; and that in the other contrast ACC is significant and IFG isn't). This dissociation was the key prediction.It is NOT the case that the authors...oh, fuck it...that I made a simple prediction of "no ACC", found "no ACC", and drew a conclusion. Instead, I predicted a specific overall pattern, a double dissociation, and this positive prediction was confirmed.The later conclusions that ACC is not differentially active derives not JUST from this test, but also the numerical data (fig 3b) AND, more importantly, *other reports*.If a hypothesis predicts "no difference", it's true that standard statistical tests don't exist to "confirm" this non-difference. However, this hypothesis does make a prediction. Furthermore, this prediction can be falsified (by finding a significant difference). If multiple failures to falsify the hypothesis occur, then one's confidence in the hypothesis increases.This is a thorny area, but this is not the type of thinking you classified it as. If I'm making a logical error, it's a more subtle one! What to do with predictions of "no difference" is a thorny issue, but what I outlined above - finding patterns of activation, not just a single failure, and relying on multiple replications in the literaure - is a reasonable and commonly held line of reasoning.And, finally - you are right sometimes. I have read people making very bold, innappropriate claims regarding non-significant differences. I have, I should say, seen this in biological papers, too, not just psychological papers.

Trust me, studies of some part of the brain can be very bad science, or pseudoscience.It is not WHAT is being studied, but HOW.I gave an abstract of a paper in which the only data being analyzed was the subjective ratings of "trustworthiness" and "approachabitly" of people in pictures or of people described in written (or verbal) descriptions. The groups of subjects were "controls" and people with brain damage to the amygdala.You backed off of the strong version of your claim when presented with this abstract. I'm assuming you consider that paper "science", although you didn't comment in detail, so correct me if I'm wrong.What if the groups of people were "controls" and "people with a history of sexual abuse". Or controls and "people who have been diagnosed with anxiety disorder". Or controls and, well, whatever.Would that still be science? Note that the methods and analyses and type of data would still be the same.

Re: Evo-Psych.I haven't read the article in question, so I won't comment on its quality. I only included it as an example of biological explanations within psychology, as a counter to the claim that psychology is hostile to biological explanations. This is indeed a sad state of affairs, but the blame here must rest primarily on an increasingly theocratic government threatening these groups. This was a strategic move on their part to prevent drastic governmental oversight and interference. Better to have a toothless policy yourself than to have an ignorant Congress institute oversight themselves. I'm not happy about any of this, but it has far more to do with politics than the status of experimental psychology as science. It isn't "unthinkable" for me. It wouldn't happen right here, right now. But similar situations have happened in the past, and I, for one, won't rule it out for the future, either.