Jessica Turner, Ph.D. Mind Research Network, Albuquerque, NM Angela Laird, Ph.D. University of Texas Health Sciences Center, San Antonio 1.

1

What is the relationship between cognitive processes and cognitive

experiments?

Jessica Turner, Ph.D.Mind Research Network, Albuquerque, NM

Angela Laird, Ph.D.University of Texas Health Sciences Center, San Antonio

2

ConclusionsMental functions may be shorthand for complex

neural circuitry and function; or they may be emergent properties which cannot be

further deconstructedEither way we can build an ontology for them

The link between experimental results and the mental function they claim to be “about” needs to be clarifiedNeed details of time, location, assumptions, methods.Need to consider the hypothetical framework; the

explicit operationalizations; the analyses done and not done; and the caveats from the experimental context.

3

Core Symptom Clusters in Schizophrenia: Or, why I care about mental function

DelusionsHallucinationsDisorganization

I. Positive symptomsBlunted affect

Few words (Alogia )No initiative (Avolition )No pleasure (Anhedonia)

II. Negative symptoms

Discontent/depression (Dysphoria )Suicidality

Hopelessness

IV. Affective symptoms

Social/occupational dysfunctionWork/interpersonal relationships

Self-care

AttentionMemory

Executive functions(eg, abstraction)

III. Cognitive symptoms

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. Fourth ed. Text Revision. Washington DC: American Psychiatric Association. 2000.

Slide courtesy of Dr. Jose Cañive

4

The ontological status of mental function over timeIntrospectionism

Behaviorism

Cognitive science

5

In the extremes

Eliminative materialism

Mental functions as concepts will disappear as we understand the brain better

Reality is epiphenomena

All we have is consciousness and everything else is a model

6

Cognitive Paradigm Approach

The mental function experimenters claim to be studying

is not as important as how they study it.

7

Schema for cognitive experimentsBrainMap database: Experiments have

conditions

Describe the conditions of the experimentStimulusInstructionsResponse

And context: Pre/post treatment? Population studied?

And behavioral domain

8

CogPO basics

Turner and Laird, 2012.

9

BrainMap behavioral domains

CogAtlas is similarNot completeNot clear which is whichNot orthogonal

10

Linking mind and behaviorCognitive experiments attempt to measure

mental processes

Donders, 1869.

11

SZ have deficits detecting low spatial frequencies

O’Donnell et al., J. Abnormal Psychology 2002.

12

Disorganized perception

Louis Wain (1860-1939)

13

Sz have attention problems

Controls on average: 77% correctSz on average:

54% correct

Carter et al., Schizophrenia Res. 2010

Disentangling the roles of attention and perception in measuring perceptual thresholds is not easy to do.

14

Linking mind and physiologyCognitive neuroscience experiments attempt to measure

physiological correlates of behavior, indirectly connected to mental processes

Attention: What is it? Endogenous (voluntary) attention vs Exogenous (involuntary)

attention Posner: Experimental designs for studying voluntary attention

The differences in speed and accuracy among these trials relates to voluntary control of attention

The spatiotemporal receptive-field map of a single neuron in the unattended mode (top) and attended mode (bottom).

McAdams C J , Reid R C J. Neurosci. 2005;25:11023-11033

©2005 by Society for Neuroscience

16

Cognitive neuroimaging and mental function

Localization of function

Effort-related increases in signal

17

Cognitive neuroimaging and mental function

Carter et al., Schizophrenia Res. 2010

Controls (L) and SZ (right) increases in BOLD signal for different attentional task conditions.

18

Linking mind and brain Brain Map-supported analysis of “executive function” in

schizophrenia Also known as “cognitive control” 41 papers with tasks including delayed match-to-sample or delayed

response (including Sternberg item recognition), go/no-go (including AX-CPT), mental arithmetic, N-back, oddball, sequence recall, Stroop, Wisconsin Card Sort, and word generation tasks

Minzenberg et al., Arch. Gen. Psychiatry, 2009

19

Linking mind and brain

20

From experiment to mental functionHypothesesOperationalizationAnalysisInterpretation

21

HypothesesWe frame hypotheses within a scientific

framework.

This rules out some questions as being sillyE.g. we don’t ask about the olfactory role of

primary visual areas

But also keeps us from asking other questionsCan we treat schizophrenia with cognitive

training?

22

OperationalizationThis is a key step in linking what we are studying in the

abstract to what we are measuring in reality

Attention: Performance differences somehow capture what we mean by

attention

Perception:Sensory thresholds are defined as the 50% point or 75% point

Executive functionHow many trials do you perseverate in using a rule you’re

being told is wrong?

23

AnalysisWhat we do with

data is shaped by what we think is acceptable within our framework

24

InterpretationThe caveats are important!These are plausible alternative explanations

for the experimentFailures of operationalization

Eg. A monkey who has figured out another way to get their juice reward or a human who is doing the study in some completely unexpected way

The BOLD signal or other measurement is not sensitive to the differences you want

Or it is sensitive to differences you don’t want (medication)

25

Cognitive neuroimaging and mental functionExamples from fMRI

It gets messy, when the linking hypotheses aren’t clear

E.g. clinical populations: We know that the BOLD signal doesn’t look the same, and the behavior is different; but what does that mean about mental function per se?

26

ConclusionsScience does not work solely by building on logical

axioms. There is a theoretical framework behind the

formation and interpretation of every experiment.How we think about the experimental design and

interpretation is not context-free. Over time, concepts and relationships will disappear

and new ones show up E.g., phlogiston, DNA as a blue print, arguments over

nature vs nurture, the role of the unconscious in mental dysfunction.

Our semantic framework for reasoning within cognitive neuroscience has to take that into account.

27

ConclusionsHow we characterize mental functions is

going to change (again!).

The results of cognitive experiments about mental function need to be subscripted by the experimental methods and theoretical framework.