Interplay between cognitive & biological psychology to impact behaviour

The scientific stufy of biology of behaviour (Dewsbury, 1991)

• Denotes a biological approach to the study of psychology. 
• Psychology is the scientifc study of behaviour; the scientific study of all overt activities of the organism as well as the internal processes that are presumed to underlie them.

Neuroanatomy: The study of the structure of the nervous system.

Neurochemistry: The study of the chemical bases of neural activity.

Neuroendocrinology: The study of interactions between the nervous system and the endocrine system.

Neuropathology: The study of nervous system dysfunction.

Neurophramacology: The study of the effects of drugs on neural activity.

Neurophysiology: The study of the functions and activities of the nervous system.

Example Korsakoff's syndrome 

• Good sensory, motor, and cognitive abilities, but forget within a few seconds (hence confabulation).
• Neuropsychology:
• Patients with lesions (especially in thalamus, hypothalamus, and mammillary bodies, mainly due to alcoholism leading to thiamine deficiency).
• Comparative Psychology:
• Thiamine-deficient rats displayed similar behaviour to anatomical lesions.

• Physiological psychology:
• Experiments on rats with damage to hippocampus.
• Psychophrarmacology:
• Pharmaceutical approaches (attempts) aimed at acetylcholine receptors.
• Neuropsychology:
• E.g., Wernicke-Korsakoff's syndrome discussed above.
• Psychophysiology:
• Galvanic skin response and other physiological responses to a face can occur even when a patient does not recognise the face.
• Cognitive Neuroscience:
• fMRI experiments to show changes in parts of the brain while performing memory tasks.
• Comparative Psychology:
• Birds who hide seeds have a relatively large hippocampus.

Pinel, p33-36

Concerned with the internal processes involved in making sense of the environment and deciding on appropriate action.

• Processes include: attention, perception, learning, memory, language, problem solving, reasoning and thinking. 

Definition: Aiming to understand human cognition by observing their behaviour of people performing various cognitive tasks.

broad definition: Brain activity and structure as relevant information for understanding human conditions.

Cognitive Psychology:

• Uses behavioural evidence to help understand human cognition.
• Limitation: models have often been described verbally (rather than as a computational model).
• Limitation: Sometimes a proliferation of theories (e.g., of working memory).
• Limitation: Sometimes over-specific (applies only to a limited set of paradigms).
• Limitation: "Hard to identify the number and nature of processes involved on the basis of speed and accuracy measures" (Eysenck & Keane, 2020, Chapter 1, p7). Prone to boxology.

• Uses pattern of performance of brain-damaged patients on cognitive tasks to make inferences about the mechanisms of cognition in healthy individuals.
• Assumption: modularity of processes.
• Assumption: Anatomical modulaity (not necessarily the same as modularity of processes).
• Assumption: subtractivity (damange impairs one thing but leaves others intact) and transparency (removing a modulle from a model should have the same effet as the damage in a patient).
• Given these assumptions, double dissociations should be strong evidence for a 'module' (one patient performs normally on task X and is impaired on task Y but another patient shows the opposite pattern).
• There has been progress in combining cognitive neuropsychology with cognitive neuroscience to relate cognitive medols and anatomical regions of the brain.

Limitation: The idea of modularity is being questioned. e.g., artificial neural networks are not modular. This is threatening to the logic of neuropsychology. Effects of a lesion can be widespread, affecting distant but connected parts of the brain.

Limitation: Compensation occurs (patients recover function to some extent, i.e., plasticity).

Involves "intensive study of brain activity as well as behaviour". (So cognitive and biological psychology).

Techniques:

• Single-unit recording: invasive techniques.
• ERP, event-related potentials (EEG locked to stimulus onset).
• PET, positron emission tomography (rare, but earlier than fMRI. Used in early studies, e.g., Zeki: V4 vs MT)
• fMRI
• MEG, magneto-encephalography. Poorer spatial resolution than fMRI but much better temporal resolution.
• TMS, transcranial magnetic stimulation. Temporary disruption to localised regions of cortex.
• tDCS, transcranial direct current stimulation.

Success: Neural decoding

• E.g., lots of images > record fMRI > calcuate function relating images to response > predict for novel stimuli
• sometimes this technique performs better than the observer.
• Success: Resolve debates, e.g., is visual imagery & visual stimulation similar?
• Very similar activations in visual cortex 
• Success: Coordinated data analysis
• Sharing of raw data for re-analysis
• Meta-analyses for more generalisable conclusions.
• Success: TMS and timing
• Improved understanding of when activity is critical 
• Limitation: 'blobology', e.g., fusiform face area (FFA) 'is responsible for' face recognition.
• Limitation: usually studies report only the localisation of cortical processing associated with a task (9/10), not testing a cognitive theory (1/10, Tressoldi et al., 2012). This may have improved since then.

Eysenck & Keane, p16-25

• Progress?: MVPA (multi-voxel pattern analysis)
• Does not assume that all the processing associated with one process/task is situated in one place (a 'blob' on fMRI scan), so a complex pattern of different voxels can take the place of a 'blob' e.g. Mahmoudi et al., 2012)
• Reliable?: A study askign 70 groups to re-analyse a set of fMRI data found an alarmingly wider range of interpretations of the same data (Botvinik-Nezer et al., 2020)
• "Here we assess ... [the variability caused by] ... asking 70 independent teams to analyse the same dataset ... [This showed] ... an overestimation of the likelihood of significant findings ... [The analysis] revealed substantial variability in reported binary results, with high levels of disagreement across teams for most of the tested hypotheses."

• Computational modelling involves programming computers to model or mimic human cognitive functioning. If we can mimic both the sucesses and the failures of a human carrying out a cognitive taks, it will improve our understanding of the processes the brain carries out.
• Review by Kriegeskorte & Douglas (2018):
• "To link mind and brain, models must attempt to capture aspects of both behaviour and neuronal dynamics".