Welcome to Last Minute Lecture.
This free chapter overview is designed to help students review and understand key concepts.
These summaries supplement not replaced the original textbook and may not be redistributed or resold.
For complete coverage, always consult the official text.
If you break a bone,
an x -ray gives you this clean jagged white line.
It is completely binary.
Right, it's broken or it's not.
Exactly.
The doctor points to the film and the diagnosis is just right there in high contrast.
But how do you take an x -ray of a working memory or a fleeting thought?
Yeah, that is the ultimate diagnostic challenge.
It really is.
Welcome to the deep dive, by the way.
We've crafted this specifically for you, the Last Minute Lecture team.
Consider us your personal one -on -one tutors for Chapter 13 of Introduction to Neuropsychology Second Edition.
Which is all about electrophysiology and imaging.
And it's funny because about 25 years ago, electrophysiology was supposed to be this magical bridge.
Right, the bridge that would finally solve the mind -body problem.
Exactly.
The thought was, hey, if we can just record the brain's electricity, we can watch mental events happen in real time.
But then CT and fMRI scans came along and kind of stole the spotlight for research.
They did, yeah.
But electrophysiology is still absolutely critical clinically, especially for things like seizure disorders.
It's still a really vital, cost -effective tool.
So our mission today is to walk you through this material in the exact order of the text.
We are going from, you know, basic electrical recordings all the way up to the most cutting edge brain scanners.
Right.
And to understand the modern high -definition stuff, we really have to start with the baseline, which is the ongoing EEG.
The electroencephalogram.
That's the one.
It's been around since the 1930s.
Okay.
Let's unpack this.
How does it actually work?
Like the mechanics of hooking up the wires?
Well, you basically use a specialized cap to place these small silver cups on the scalp.
Which you can actually see in figure 13 .1, right?
That classic picture of the moving paper chart showing bursts of alpha activity.
Yeah, exactly.
And the placement isn't random.
You follow what's called the 10 -20 system.
That's figure 13 .2 for anyone following along in the book.
Right.
You're spacing the electrodes at 10 % and 20 % intervals between these fixed anatomical landmarks on the skull.
But you can't just put the cups on dry hair, right?
You need gel.
Oh, absolutely.
You have to fill them with a highly conductive gel.
You're essentially filling in all the microscopic potholes on the skin so the signal doesn't hit air pockets.
Because the brain's electrical fluctuations are tiny.
Incredibly tiny.
We're talking microvolts.
They have to be amplified like 20 ,000 times just to create a readable trace.
Wow.
Okay, but wait.
That brings up the whole reference -electra dilemma?
Yes, it does.
Because if you're doing a bipolar recording where you measure the difference between two active spots on the head, how do you actually know which spot changed?
It's like trying to eavesdrop on a single conversation through a thick wall at a noisy party.
That is a perfect analogy.
Right.
If the volume spikes, you don't know who just yelled.
Exactly.
So the technical workaround is what we call a monocular recording.
You try to find a relatively inactive reference spot to compare your active electrode against.
Like where?
The nose?
The nose?
Actually, yes.
Or the vertex, which is the midpoint on top of the head.
Or the mastoid's behind the ears.
But no spot is totally neutral.
No, not at all.
The anatomical asymmetries in the brain mean there is no perfectly neutral ground.
A truly unbiased measurement is basically impossible.
Right.
Okay, so once the electrodes are glued on and the data starts flowing,
you're getting, what is it,
61 ,440 numbers per minute across just eight channels.
There's something staggering like that, yeah.
How do researchers even make sense of those squiggles?
Well, you break them down into frequency bands.
Right, measured in hertz.
So we've got delta, which is 0 to 3 .5 hertz, theta is 4 to 7 .5, alpha is 8 to 12 .5, and beta 1 is 13 to 19 .5, and beta 2 is 20 to 29 .5.
Exactly.
And the key thing to look for is what we call alpha attenuation.
Which is when alpha power drops.
Right.
So when you're just resting, your neurons are sort of idling together.
They fire and sink, which creates that alpha wave.
But when you start a mental task, that alpha power drops and beta activity shoots up.
Because the neurons are breaking formation.
They have to do independent work.
Precisely.
But here is the issue.
Just measuring the overall power of a frequency band is pretty crude.
It's like trying to judge an orchestra just by measuring how loud the violins are.
Yes.
It completely misses how the instruments are playing together.
Exactly.
And what's fascinating here is that researchers realized this and developed coherence analysis.
Oh, okay.
How does that work?
So coherence doesn't just measure the volume or power.
It measures the shared, synchronized activity between two different channels.
And it does this completely independent of power.
So it's building a dynamic map of how regions talk to each other.
Exactly.
Down to a half -second epoch.
It lets researchers see cognitive events happening in real time.
Okay, so ongoing EEG is great for general states.
But what if we want to isolate the brain's exact reaction to one specific thing, like hearing a single tone?
Then you need evoked potentials.
Or EPs.
Right.
And this is figure 13 .3 in the text.
This relies on computer averaging, right?
It does.
You present the exact same stimulus, like that tone,
maybe 64 to 512 times.
And because the background noise of the brain is random, it fluctuates up and down.
Exactly.
So mathematically, if you average all those hundreds of recordings together, the random noise cancels out.
Leaving just the true underlying waveform.
And figure 13 .4 shows how those wave shapes look totally different for visual, auditory, and somatosensory EPs.
Right.
The shapes are very distinct.
But wait, I have a major issue with this.
Doesn't this assume the brain's reaction time is identical every single time?
Ah, yes.
You're talking about the latency jitter problem.
Yeah, figure 13 .5.
If the brain is even a few milliseconds late reacting to one of the tones, the average gets totally smeared.
It's like a choir singing the same note.
Right.
If everyone enters at slightly different times, the sharp attack of the note just gets flattened.
That's exactly what happens.
Researchers have to use complex filters to align those peaks before they average them.
And once they get a clean wave, they classify the peaks.
Right.
They classify them into components.
You have exogenous components, which are really early, early under 100 milliseconds, like the N1 wave.
And those are just raw sensory reception.
Yep.
Purely driven by the physical stimulus.
But then you have endogenous components.
Those are middle to late, between 100 and 1 ,000 milliseconds.
Like the famous P300 wave.
Exactly.
And those reflect cognitive evaluation and decision making.
It's the brain actually thinking about the stimulus.
So armed with EEG and EPs, researchers tried to test how the left and right hemispheres divide cognitive labor.
They did.
And the results were, frankly, a total mess.
Yeah.
Because you had Davidson and Ehrlichman finding the expected alpha drops in engaged hemispheres.
But then Beaumont and Rugg and Ornstein found the exact opposite.
They found alpha enhancement or just no effects at all.
Which is incredibly frustrating if you're trying to build a consistent theory.
Here's where it gets really interesting, though.
Were they actually measuring brain power or just the physical effort of the task, like eye twitches or something?
That was the big question.
And Gevens ran some famous studies to test that.
Oh, right.
When he strictly controlled variables like eye movements, limb movements, and overall effort,
all those task -related EEG asymmetries completely vanished.
Wow.
So the textbook lateralization stuff was just artifacts.
A lot of it was, yeah.
But to be fair, the EP studies were a bit more successful.
Because they're locked to a specific time.
Exactly.
Wood found actual left hemisphere phonetic processing differences.
And Brown and Marsh did this really cool study.
That's the ambiguous words study.
Yes.
They showed that anterior cerebral processing was different when people heard sentences like the metal was lead versus the horse was lead.
That's amazing.
But still, electrophysiology was plagued by these artifacts.
And the spatial resolution was terrible.
You knew when something happened, but not where.
Right.
So in the 1970s, we see this massive pivot towards structural imaging, starting with the CT scan.
Computerized chymography.
Hans Fehl and Cormac actually won a Nobel Prize for it.
Because they figured out how to use X -rays from multiple angles to construct these visual slices of the brain,
where bone is white, air and water are dark, and the brain itself is gray.
Exactly.
It was revolutionary.
But compared to today, it's kind of like playing an old 8 -bit video game.
I mean, it's great for spotting gross trauma like a bleed, but terrible for nuance.
How did we upgrade to high definition?
That was magnetic resonance imaging, or MRI.
No X -rays this time, right?
Right.
No radiation.
You place the body in a powerful magnetic field, which aligns the atomic particles.
Then you hit them with radio waves.
And the way those particles relax creates the image.
Exactly.
It gives gorgeous gold standard anatomical clarity, but there's a huge catch.
It only shows structure.
Right.
It shows you the engine, but not the engine running.
So to see the engine running, we got fMRI, functional MRI.
Yes.
Which tracks blood glucose and oxygen going to active neurons.
It measures it multiple times a second.
Using the subtraction technique, right?
That's so clever.
Subtracting a resting state scan from a task state scan so you only see what specifically lights up.
It's brilliant.
Like Bornaz's study, they used fMRI to prove that answering questions about how tools are manipulated specifically activates the left inferior parietal lobe.
Which is incredible.
But wait, what about PET scans, positron emission tomography?
Oh, PET is older than fMRI.
It uses radioactive isotopes instead of tracking oxygen.
But it's super slow, right?
Like 30 seconds per scan.
It is slow.
But it yielded some brilliant results.
McGuire's same as study on London taxi drivers navigating the knowledge use PET.
And that showed the right hippocampus lighting up.
I have a highly practical question, though, about fMRI.
How valid are these cognitive tests when you shove a subject inside a noisy, claustrophobic, terrifying magnetic tube?
This raises an important question, honestly.
It's a huge stressor.
Right.
And Gucheson Park actually found that the stressful fMRI environment inherently alters cognition.
It can actively impair a subject's memory performance during the test.
So you're measuring a stressed brain, not a normal one.
Exactly.
So if fMRI is slow and claustrophobic and EEG is just squiggles,
what is the ultimate future of brain measurement?
That would be MEG.
Magnetoencephalography.
I love MEG.
It detects the brain's magnetic signals, right?
Yes.
But those signals are incredibly weak.
To detect them, you need superconducting coils bathed in liquid helium.
At minus 269 degrees Celsius?
That is wild.
It's extreme engineering.
The sensitivity is like trying to hear the footsteps of an ant in the middle of a rock concert.
Wow.
But the payoff is insane.
A spatial resolution of two millimeters and a temporal resolution of one millisecond.
It's the best of both worlds.
Kay Conan actually used MEG to show exactly how alcohol impairs auditory processing millisecond by millisecond.
Okay, one final structural innovation we have to mention.
DTI.
Diffusion tensor imaging.
Or fiber tractography.
Ah, yes.
This is basically mapping the brain's highway system, right?
By tracking how water diffuses along the white matter track.
Exactly.
Because water diffuses differently alongside those long neural cables, Lee used it to understand the physical wiring behind blindsight.
Cortically blind patients who can react to objects they can't consciously see.
Exactly.
It maps those subconscious pathways.
Okay.
So to synthesize all this for the last minute lecture team,
brain imaging is incredibly fashionable right now.
It is.
But we have to remember, it is historically over -promised.
Right.
We thought EDG is the answer, then we thought fMRI was the answer.
And the real breakthroughs of the future won't come from just building bigger magnets.
It's going to come from combining these glossy imaging techniques with traditional clinical lesion studies.
So they mutually inform each other.
Exactly.
Well, I want to leave you with a final provocative thought to mull over.
If we are getting to the point where we can map the exact white matter highways with DTI and track cognition millisecond by millisecond with MEG, how far are we from being able to read a thought before the person even realizes they are thinking it?
It's a slightly terrifying but fascinating question.
It really is.
A huge thank you specifically to the last minute lecture team for learning alongside us today.
We really hope this one -on -one deep dive helped you master Chapter 13.
Best of luck on your exam.
You're going to do great.
You've got this.
Thanks for tuning in.