Deck 3: Recognizing Objects

Question 1

You are shown an odd-looking image and asked to identify it. According to our knowledge of object recognition, your first step would be gathering the raw data, and the second would be

A) memorizing the data.
B) ignoring the irrelevant data.
C) interpreting the data.
D) suppressing the data.

Answer 1

C

Question 2

Researchers have used transcranial magnetic stimulation (TMS) to disrupt portions of the healthy brain. When asking participants to search for a target, we expect TMS applied to the parietal lobe to

A) have no effect on participants who have not suffered parietal damage.
B) disrupt the search for both a target defined by a single feature (e.g., "Find the red shape") and a target defined by a conjunction of features (e.g., "Find the shape that is red and round").
C) disrupt the search for a target defined by a single feature (e.g., "Find the red shape").
D) disrupt the search for a target defined by a conjunction of features (e.g., "Find the shape that is red and round").

Answer 2

D

Question 3

If visual information conflicts with other information, we usually

A) trust the visual input.
B) distrust the visual input, as our eyes can play tricks on us.
C) trust input from other senses, like auditory input.
D) become confused and give up.

Answer 3

A

Question 4

The importance of vision for humans is reflected in the

A) close proximity of the eyes to the visual cortex.
B) inability of brain damage to disrupt the visual system.
C) lack of a "blind spot" in humans.
D) relative size of the visual cortex.

Question 5

One might propose that pattern recognition begins with the identification of simple elements; these are then added together, like tiles within a mosaic, to allow recognition of the large-scale pattern. Which of the following statements is MOST problematic for this proposal?

A) The elements identified within a pattern can be changed dramatically by a change in context.
B) Patterns can be identified even when viewed from a novel distance.
C) Many variations on a single pattern can be recognized, despite differences from one variation to the next.
D) Patterns can be identified even when viewed from a novel orientation.

Question 6

What sort of processing is dependent on factors in the environment or in the stimulus?

A) top-down
B) bottom-up
C) expectation-based
D) knowledge-driven

Question 7

In a tachistoscopic procedure, a word is likely to be more difficult to recognize if it

A) has been encountered by the participant recently.
B) is used frequently in the language.
C) has an unusual spelling pattern.
D) has been primed by an earlier exposure.

Question 8

"Bottom-up" (or "data-driven") mechanisms are

A) the scientific process in which all claims must be rooted in well-established biological evidence.
B) mechanisms for which activity is primarily triggered and shaped by the incoming stimulus information.
C) mechanisms for which activity is influenced by thoughts provided by the individual.
D) the process by which researchers seek to develop new theories by paying close attention to the available data.

Question 9

Which of the following statements is NOT true for feature-based models of pattern recognition?

A) Features, as general-purpose building blocks, can help explain how humans recognize variations on a form (e.g., a cat in different positions or a letter in different type fonts).
B) The visual system identifies small pieces of a pattern first and then combines them to form more complex wholes.
C) Studies using people with parietal damage suggest that the detection of features is separate from the process to put the features together.
D) The term "features" can also be used to describe a distinct object made up of several parts.

Question 10

Participants are shown a visual stimulus for just 30 ms and are then asked, "Was there an E or a K in the stimulus?" We would expect the BEST performance if the stimulus is

A) BARK.
B) K.
C) BWQK.
D) GALK.

Question 11

Imagine you are reading a puzzling email from a friend. You identify the words, but have a hard time "reading between the lines." In this example, word identification involves _____ processing while "reading between the lines" involves _____ processing.

A) bottom-up; top-down
B) top-down; bottom-up
C) bottom-up; bottom-up
D) top-down; top-down

Question 12

Imagine you are putting together a puzzle. The lid of the box comes with a picture of the completed puzzle, and you reference that while you are working. The lid is acting as a

A) top-down influence.
B) bottom-up influence.
C) expectation.
D) distraction.

Question 13

Which of the following sentences best illustrates the effect that Gestalt principles have on perception?

A) "Go beyond the information given."
B) "Seeing is believing."
C) "Think outside the box."
D) "Believing is seeing."

Question 14

In tachistoscopic studies, a poststimulus mask is usually employed to

A) disrupt sensory processing of the stimulus.
B) prevent verbalization.
C) help the participants maintain proper eye position.
D) discourage guessing about the stimulus.

Question 15

The Necker cube is an example of an ambiguous figure. Which of the following statements regarding the cube is NOT true?

A) The information given in the drawing does not cause perception of the cube in one orientation over the other.
B) The lines on the page are neutral in regard to the shape's configuration in depth.
C) The lines on the page contradict one another and so a cube can never be perceived.
D) There is more than one perceptual interpretation of the cube's orientation.

Question 16

The "word-superiority effect" refers to the fact that it is easier to recognize

A) short (three- or four-letter) words than long words.
B) a letter within the context of a word than it is to recognize a letter presented by itself.
C) a word presented in a phrase than it is to recognize a word presented by itself.
D) words that are frequently used under tachistoscopic conditions.

Question 17

It is suggested that features have special status. Which of the following findings does NOT support this hypothesis?

A) Figures with single features are detected very easily in visual search tasks.
B) People with integrative agnosia can detect features but cannot combine them.
C) Feature recognition is separate and occurs before recognition of objects.
D) Perception of features changes based on the perceiver's expectations but perception of objects does not.

Question 18

In order to summarize the Gestalt psychologists' movement in a few words, one might say,

A) "If you can't see it happen, it isn't worth studying."
B) "The perceptual whole is different than the sum of its parts."
C) "All that is important happens in the subconscious."
D) "What you see is what you get."

Question 19

A participant reads a list of words in which the word "elephant" appears several times. Later, the participant tachistoscopically views another list of words. When the word "elephant" appears in the second list, the participant's response rate is faster than for other words not found on the previous list. This effect is called

A) the word-superiority effect.
B) the redundancy claim.
C) working-memory availability.
D) repetition priming.

Question 20

A tachistoscope is a device used to

A) measure the rate at which a neuron is firing.
B) provide precise measurements of reaction times.
C) display stimuli briefly.
D) record the moment-by-moment activities of the brain.

Question 21

On one trial of an experiment, a participant is shown the sequence GWXT. On a different trial, the participant is shown the sequence PAFE. On the basis of prior research, we should expect that

A) PAFE will be easier to perceive than GWXT because detectors for PA and FE are likely to be well primed.
B) the letter sequences will be equally difficult to perceive because neither is a word.
C) participants will perceive more of the letters in GWXT because they are likely to confuse PAFE with PACE or SAFE.
D) the letter sequences will be equally difficult to perceive because both contain regular bigram patterns.

Question 22

A feature net is a

A) network of cognitive "detectors" organized in hierarchical layers.
B) collective of features used to describe an object's form.
C) netlike structure of brain cells designed to detect features.
D) conceptualization of how features are visually related to one another.

Question 23

A bigram detector fires in response to the

A) appropriate object weight.
B) appropriately shaped curve.
C) appropriately positioned corner.
D) appropriate letter pair.

Question 24

In a feature-net model, knowledge of spelling patterns

A) can influence the perception of whole words but not the perception of single letters or bigrams.
B) is distributed across the model, and therefore the knowledge is only detectable in the overall functioning of the network.
C) is locally represented, allowing the network to draw inferences about partially viewed stimuli.
D) is overshadowed by the parallel processing employed by the net.

Question 25

Mistakes in word reading occur under a feature-net model of recognition. This results because the feature net encourages _____ over _____.

A) accuracy; efficiency
B) efficiency; accuracy
C) laziness; hard work
D) bottom-up processing; top-down processing

Question 26

A response threshold is the

A) exposure duration for which a word must be displayed tachistoscopically for a particular participant to perceive it.
B) number of correct responses required in order for a participant to perform above average on a particular task.
C) amount of certainty or conviction a participant expresses when selecting a particular response.
D) activation level at which a response occurs.

Question 27

We often encounter ambiguous letters when reading handwritten words but can still interpret them. For example, the same shape can be interpreted as an A in CAT but an H in THE. At what level of analysis does the feature net resolve this issue?

A) the bigram level
B) the letter level
C) the word level
D) overregularization

Question 28

Participants in a tachistoscopic procedure are shown the sequence CQRN. Participants misperceive this string as CORN. In a feature-net account, which of the following statements probably does NOT contribute to this effect?

A) O is a more frequently used letter in English than is Q. Therefore, the O-detector is better primed.
B) CO is a more frequent letter pair in English than CQ. Therefore, the CO-detector is better primed.
C) A well-primed bigram detector will fire even if the letter detectors feeding into that bigram detector are firing weakly.
D) Feature nets are generally unable to identify nonwords.

Question 29

One type of error that can result from feature nets is overregularization. Is overregularization a significant problem?

A) Yes; it leads to many errors.
B) Yes; not many errors occur, but they are really devastating.
C) No; these errors are infrequent and usually not problematic.
D) No; these errors occur often, but are small and easily corrected.

Question 30

In tachistoscopic recognition, participants often make overregularization errors. These are errors in which participants

A) perceive a word as pertaining to their personal experiences even when the word is relatively neutral.
B) perceive a word as being related to the previous word when in fact it is not.
C) are shown a frequently used word but perceive it as an infrequently used word.
D) are shown a pattern such as MJLK but perceive it as MILK.

Question 31

Compared to detectors that have not fired recently, a detector that HAS fired recently is likely to

A) be at a higher position within the network of detectors.
B) have a higher response threshold.
C) have a higher activation level.
D) require more priming in order to fire.

Question 32

The efficiency of a feature-net model occasionally leads to errors. We could avoid making word-recognition mistakes if we scrutinized each letter. Why is this a suboptimal strategy (and not the one we use)?

A) Reading would be very slow.
B) We do not have the attentional resources to scrutinize every letter.
C) The visual system would be overwhelmed.
D) Letters are easy to discriminate from one another.

Question 33

English nonwords (e.g., "HICE") are easier to perceive than strings of letters not resembling English words (e.g., "RSFK") because

A) they are encountered more often.
B) bigram detectors for more common letter combinations fire more readily.
C) they are more distinctive.
D) word detectors will respond to near-words as well as true words.

Question 34

When identifying nonword letter-strings that are presented very briefly, participants tend to make specific kinds of errors. How would these errors be best described?

A) They are unable to identify any letters if the string is a nonword.
B) They identify many of the letters correctly but tend to incorrectly identify the vowels.
C) They tend to misidentify strange letter combinations as more common letter combinations.
D) They misidentify more common letter combinations as less common letter pairs.

Question 35

This chapter describes in detail one way a feature net can be designed, but other designs may turn out to be preferable. For example, more recent hypotheses, including McClelland and Rumelhart's model, make use of all of the following statements EXCEPT

A) inhibitory connections among the detectors.
B) the elimination of feature detectors, relying instead on geon detectors.
C) connections allowing detectors at one level in the network to influence detectors at lower levels.
D) connections allowing detectors at one level in the network to influence other detectors at the same level.

Question 36

Participants' recognition thresholds are

A) lower for frequently seen words.
B) higher for recently seen words.
C) not affected by priming.
D) lower for highly unusual words.

Question 37

Participants are shown the letter-string TPUM for 30 ms and asked to identify what they saw. If they are going to answer incorrectly, which response are they most likely to give?

A) I did not see anything presented.
B) OPUM
C) TRUM
D) TMPU

Question 38

In a tachistoscopic procedure, participants are shown the sequence NACL. Evidence indicates that

A) the distinctive letter pattern in the sequence will help participants recognize the sequence.
B) the familiarity of the sequence (i.e., the chemical formula for table salt) will help participants recognize the sequence.
C) participants are likely to misperceive the sequence, reading it as if it were a common letter pattern, such as NAIL.
D) participants will be unable to organize the letters, and therefore they will only perceive some of the sequence's features, not the large-scale units.

Question 39

Recognition errors (like mistaking CQRN for CORN ) lead us to what conclusion about feature nets?

A) Feature nets are an imperfect system, and thus unlikely to accurately represent our cognitive processing.
B) The interactive nature of feature nets usually allows us to identify stimuli, but can also lead to errors.
C) Feature nets are capable of explaining accurate performance, but not errors.
D) Feature nets are the only possible explanation for object recognition.

Question 40

The bigram CO is more common in English than CQ, which makes identifying CORN more likely than identifying CQRN. According to feature nets, how is knowledge of bigram frequency stored?

A) It is locally represented in the feature net.
B) It is not explicitly stored anywhere.
C) It is built into a specific bigram-detection process.
D) It is stored in the hippocampus, a brain structure crucial for memory.

Question 41

The term "top-down processing" can be interchanged with the term

A) "concept-driven processing."
B) "stimulus-driven processing."
C) "repetition-priming processing."
D) "interactive processing."

Question 42

Argue for or against the notion that "face processing is a special case." Support your perspective by referencing behavioral or neuropsychological evidence.

Question 43

Facial recognition differs from recognition of other objects in all of the following ways EXCEPT:

A) Facial recognition depends on holistic recognition.
B) The component parts of a face have no influence on perception.
C) The configuration of the component parts of a face is critical.
D) The component parts of a face are not considered individually.

Question 44

Which of the following statements is NOT true about a feature-net model?

A) It can be applied to recognition of print.
B) It can be applied to recognition of three-dimensional objects.
C) It can be applied to recognition of faces.
D) It is only one part of a recognition framework that also includes top-down influences.

Question 45

Biederman's recognition by components (RBC) model

A) does not rely on a hierarchy of detectors.
B) makes use of geon detectors, which in turn trigger detectors for geon assemblies.
C) asserts that priming takes place primarily at levels higher than the level of geon detectors.
D) can recognize three-dimensional objects provided they are seen from the appropriate viewing angle.

Question 46

The recognition of faces

A) seems to rely on the detection of features and geons, indicating that the recognition by components model can be applied to face recognition.
B) resembles other forms of recognition in that our ability to recognize faces is relatively unimpaired by changes in viewing angle or orientation.
C) differs from other forms of recognition in that face recognition appears not to be influenced by expectation or knowledge effects.
D) is influenced by configurational factors, suggesting that a model based on feature detection will provide a poor explanation of face recognition.

Question 47

You are driving in your car listening to the radio when a new song by your favorite artist comes on. Describe how top-down and bottom-up processing both contribute to your ability to identify (either correctly or incorrectly) the lyrics of the song.

Question 48

We can often recognize an object even if some of the object's parts are hidden from view. Evidence indicates that this recognition from partial viewing will be easiest if

A) we can see enough of the object to identify some of its geons.
B) we can see at least 20% of the object's features.
C) the object's features are unfamiliar to us so there is no risk of false alarms.
D) the object does not have too many geons.

Question 49

Describe the trade-off of efficiency versus accuracy in word recognition. Make sure to reference feature nets in your answer. Why is such a trade-off necessary?

Question 50

A researcher shows a group of participants letter-strings for a brief period of time (50 ms) and asks them to identify the letters that they saw. Based on previous research, describe the pattern of performance that is expected by answering the following questions.
a. Can participants do this task?
b. If they make mistakes, what sort of mistakes do they make?
c. What do these results tells us about object recognition?

Question 51

If instructed to identify the briefly presented word "PORK," which of the following hints will provide the greatest benefit for identification?

A) The word will be presented in capital letters.
B) The word is something you can eat.
C) The word will be shown for 50 ms.
D) The word does not contain an I.

Question 52

The form of brain damage identified as prosopagnosia is primarily characterized by an inability to

A) recognize faces.
B) comprehend written text.
C) identify inverted stimuli even though perception of upright stimuli seems normal.
D) identify familiar voices.

Question 53

According to the recognition by components (RBC) model, geons are NOT

A) simple shapes.
B) viewpoint independent.
C) always easy to identify.
D) capable of identification if they are partially obstructed.

Question 54

The fusiform face area (FFA) is known to be an area that is specifically responsive to faces. Which of the following statements is also TRUE about the FFA?

A) It is primarily thought to be the area used to mentally rotate an inverted face into its upright position for further processing.
B) It only responds to famous faces (e.g., President Barack Obama).
C) It is also a crucial area in the processing of features within an object (e.g., edges, curves).
D) Tasks requiring other subtle distinctions within a category (e.g., identifying different birds or cars) also produce high levels of activation in this area.

Question 55

Which of the following statements is TRUE about the recognition of inverted faces?

A) Recognition of inverted faces is harder than for upright faces.
B) Face processing is not affected by inverting the image.
C) Inverting a nonliving object, such as a house, produces a bigger deficit in recognition than inverting a face, as we are less familiar with houses.
D) Specialist neurons in the parietal cortex rapidly restore a face to its upright position for further processing.

Question 56

Which of the following statements BEST describe viewpoint-dependent object recognition?

A) An object is memorized faster if it is upright.
B) One must match the current view of an object with a view of the object stored in memory, using the process of rotation.
C) An object will be recognized at the same speed regardless of its orientation.
D) Recognition of an object is dependent on how many geons are visible to the viewer.

Question 57

Facial recognition depends on recognition of

A) the configuration of the parts.
B) the familiarity of the individual.
C) the lighting conditions.
D) the individual features of the face.

Question 58

Top-down mechanisms suggest that

A) upright stimuli are processed faster than inverted stimuli.
B) faces are processed faster than other body parts or inanimate objects.
C) processing can be driven by knowledge and expectations.
D) incoming information about a stimulus activates feature detectors.

Question 59

McClelland and Rumelhart's model of word recognition suggests detectors on separate levels can interact in a bidirectional manner. Biological evidence _____ this notion because _____.

A) supports; visual processing is bidirectional
B) supports; there is parallel processing in the visual system
C) does not support; visual processing is an entirely bottom-up process
D) does not support; word recognition does not depend on visual processing

Question 60

Compare and contrast the models of object recognition (McClelland and Rumelhart, geons, and multiple views).