Perception reliable, automatic and not prone to mistakes

This is a quick question regarding perception. When I read Peikoff in OPAR he explains that perception is automatic and not prone to error. Error arises on the level of the conceptual. Could someone please explain this for me a bit?

I was reading through the Objectivism 101 articles and came across this statement:

Perception is automatic and causal, and therefore reliable. Mistakes only happen when your brain starts doing some analysis, and that starts when you try to recognize the object as something you know.

It seems to me that because perception is automatic and causal does not mean it is always reliable. Perhaps I just don't understand or grasp the Objectivist view at this point. Perhaps the word "reliable" ought not have been used. I think Peikoff uses the term "valid" instead which I would accept.

However, Peikoff says some other things I wonder about. I don't have my book here so I can't check but I seem to recall that he says that perception has no power to invent or distort. This is troubling somewhat after I pulled a book from the shelf I had purchased a few years ago. I never got around to reading it. The book is a series of studies on perception. I flipped through it and found an article written by Denton - it was about how perception "fills in" certain aspects of the visual field.

He explains a couple of experiments done involving the "white noise" of a TV set. The first involved putting a small dot on the screen and a rectangle next to it. A person would focus on the dot and over time the person's perceptual field would "fill in" the rectangle with white noise. When the person would look away the rectangle would exist in the perceptual field as "white noise." He goes onto explain a second experiment using color and the person who ran the experiments determined that the brain is "filling in" for color and the white noise i.e. the brain fills in not just one specific thing.

While these are interesting experiments how is this not different from hearing "white noise" where our awareness of a particular sound slowly fades away. Or, as another example, if a person owns dogs they tend to lose the smell their dog - anyone coming into the home can immediately recognize the smell.

On the other hand, if the senses or perceptions are "reliable" how exactly is this term to be understood if it appears from these experiments that out perceptions DO invent things by "filling in?"

Anyway, just wondered about this statement in the Obj 101 articles.

Thanks

Post 1

Tuesday, September 2, 2008 - 2:48pm

Hi Tim,

Objectivism accepts a strain of direct realism, the view that your mind identifies (by automatic sense-integration) entities and their attributes that are external to your mind. Representationalists on the other hand think that entities and their attributes are just mental constructs of sorts.

Objectivists do tend to maintain that percepts aren't flawed, but there is not consensus on this point or the reasoning for it. I think the better argument to support this point -- and there are numerous arguments here -- is that percepts *by definition* are congruent with the entities or attributes that are being perceived. What appear to be flawed percepts are really just percepts with some pesky non-perceptual baggage, i.e., percepts with extraneous incongruencies -- or they are just flat out non-percepts parading around like percepts. The trouble is that *perceiving* is subject to error, whilst *percepts* themselves are not.

Per the reliability of percepts, the most popualr Objectivist argument appeals to the fallacy of the "stolen concept": despite the non-perceptual baggage affixed to some percepts, percepts and perceptual awareness are nevertheless reliable, lest we would not be able to have this discussion in the first place.

Jordan

Post 2

Wednesday, September 3, 2008 - 4:01am

Objectivism accepts a strain of direct realism, the view that your mind identifies (by automatic sense-integration) entities and their attributes that are external to your mind. Representationalists on the other hand think that entities and their attributes are just mental constructs of sorts.

I think I get what you are saying but can you possibly explain this a little more. I need to try and create a mental picture. :)

The trouble is that *perceiving* is subject to error, whilst *percepts* themselves are not.

I may need a mental picture here too. There's some slicing going on that is hard to grasp.

percepts and perceptual awareness are nevertheless reliable, lest we would not be able to have this discussion in the first place.

I agree.

Post 3

Wednesday, September 3, 2008 - 10:45am

Hi Tim,

To the first issue, representationalism says we don't perceive the external world itself. Rather, we perceive mental entities that attempt to represent the external world. In other words, for representationalists, we are stuck looking at a map or a model; we never get a direct look at the external world that the map or model represents.

Direct realism on the other hand says we look at the external world directly; we are not stuck relying on some intermediate map or model.

To the second issue, under direct realism, percepts are always real external objects. But sometimes it's tough to tell whether we are dealing with a percept by itself or with something extra or altogether different. The fact that we sometimes have trouble with discerning percepts (i.e., perceiving) doesn't affect the notion that percepts are always real external objects.

I'm not sure that helped.
Jordan

Post 4

Wednesday, September 3, 2008 - 12:29pm

Tim,

When I read Peikoff in OPAR he explains that perception is automatic and not prone to error. Error arises on the level of the conceptual. Could someone please explain this for me a bit?

Take the common illusion of a stick half-submerging in water and appear as if it were bent. The stick isn't -- in reality -- bent, so we have got to explain the discrepancy of what we see with what we get. Humans have 2 broad faculties of awareness: perceptual and conceptual. The first question to ask in the stick illusion is: which faculty screwed up?

Skepticists looking at the bent-stick illusion say it's an error in sense-perception (i.e., that we didn't perceive what was really "out there"). Objectivists realize that in order for the skepticist argument to get off of the ground, then they have got to pre-conceive that there is only one entity out there to perceive (i.e., the stick), rather than 2 of them (the stick plus the water). In short, they have to "think wrong" about what perception affords us.

Perception affords us with the temporal changes and temporal continuity (or sameness) of our ambient stimulus array. Concepts allow us to make meaning out of those changes and that sameness. A stick out of water looks straight because it is. A stick submerged looks bent because of water's refraction of light. When we see the stick in the water, we're picking up the existence of 2 entities: water and stick -- and we're, essentially, picking up the data automatically and without error.

As soon as we reach the conceptual ground of understanding that we're perceiving 2 entities, then the paradox dissolves, and we understand that our error was conceptual, rather than perceptual.

Ed

Post 5

Wednesday, September 3, 2008 - 12:42pm

Tim,

I'll try to discuss the issue of dot-focus and object fill-in after more research. Here's something I can answer now:

Or, as another example, if a person owns dogs they tend to lose the smell their dog - anyone coming into the home can immediately recognize the smell.

This is explained by the fact that our olfactory receptors fatigue. I used to work with hydrogen sulfide producing bacteria. When opening the Petri dish, the first whiff was the worst. It was almost better to take a good first whiff of the stuff -- in order to fatigue your naso-receptors and allow you to work with those little critters! I've also heard of coroners who develop so much down-regulation that they can handle formaldehyde like it isn't even there.

Ed

Post 6

Wednesday, September 3, 2008 - 4:02pm

Our sensory neurons directly correlate sensation with neural pulse rates. For example a light detecting neuron in your retina will fire faster the as more light hits it. So at the most basic level, physically, our consciousness is directly aware of its surroundings by directly converting a physical input to a electrical pulse rate.

It is true that a person can imagine a different picture than what one's sensory neurons are detecting, such as while dreaming. Potentially your neural pathway from your retina to the rest of your brain could malfunction and you could loose your vision despite having good eyes, still being able to dream with vision to almost the same detail.

A person can get high on drugs, where its detectable that one's higher level conscious awareness of their senses becomes chaotic. When this happens the senses become useless (of no help to you when trying to achieve a goal). If our senses were detached from reality, then our senses would be useless to achieve goals in reality. If our senses were detached from reality then we would have no ability to predict what will likely happen in the future.
(Edited by Dean Michael Gores on 9/03, 4:23pm)

Post 7

Tuesday, September 16, 2008 - 2:20pm

Tim, you should read David Kelley's Evidence of the Senses. It addresses this issue at length. It is the most impressive work of technical philosophy in the Objectivist canon.

Post 8

Tuesday, September 16, 2008 - 3:48pm

Ted,

I have to say that book sucks somethin' awful, mostly because it's written terribly. I chock it up to seriously bad editing, since Kelley's other works (e.g., Unrugged Individualism and The Art of Reasoning) are a pleasure to read. Still, you're probably right that it is the most exhaustive look at the sense-perception issues in Objectivist circles. But if one wants an easier-to-read but much shorter discussion of those issues, I'd recommend Sciabarra's Ayn Rand: Russian Radical. It only scratches the surface of the issues, but it's so much easier to swallow.

Jordan

Post 9

Wednesday, September 17, 2008 - 4:07am

Tim, you should read David Kelley's Evidence of the Senses. It addresses this issue at length. It is the most impressive work of technical philosophy in the Objectivist canon.

I have to say that book sucks somethin' awful, mostly because it's written terribly... ...I'd recommend Sciabarra's Ayn Rand: Russian Radical. It only scratches the surface of the issues, but it's so much easier to swallow.

I actually have Kelly's book and read it years ago but I was not as familiar with Objectivism then as now - and as Jordon points out... the editing is not so good and it made it hard for me to follow. I'll pull it off the shelf and take a look at it again.

Also, I have thumbed through some of Sciabarra's book when I was in the bookstore months ago. It looked like a good read but I didn't have the money then. Perhaps I need to take another trip to the store - I have money now. :)

Thanks guys.

I haven't been able to dig into things as of late because school has started again so I spend most of my time reading my class books and modeling stuff.

----------

I just remembered. I was thinking about this the other day. I began to consider the idea of perceptions "making inferences" and started to consider that "inference" as it applies to perceptions and "inference" as applied to conceptions are two different things. Because they are different a perceptual "inference" does not undermine anything at all.

A conceptual inference is one that we are conscious of - a perceptual inference is one we are not conscious of. The underlying processes are still causal.
(Edited by Tim Scobey on 9/17, 4:14am)

Post 10

Wednesday, September 17, 2008 - 5:30am

Tim Scobey wrote:

I just remembered. I was thinking about this the other day. I began to consider the idea of perceptions "making inferences" and started to consider that "inference" as it applies to perceptions and "inference" as applied to conceptions are two different things. Because they are different a perceptual "inference" does not undermine anything at all.

A conceptual inference is one that we are conscious of - a perceptual inference is one we are not conscious of. The underlying processes are still causal.

As I recall, this is a prominent argument in Evidence of the Senses.

Post 11

Thursday, September 18, 2008 - 5:59pm