Test questions

Question 1

TMS experiments show that the perception of one specific type of phonetic feature is influenced by stimulation of the motor cortex. Which?

a. Place of articulation
b. [+ nasal]
c. Sonority
d. Voicing

Answer 1

a. Place of articulation

Question 2

What is the action compatibility effect, ACE?

a. The increase in accuracies in judging sentences preceded by the actions these sentences describe.
b. The semantic priming effect observed between semantically congruous words, sentences and larger pieces of discourse which is manifest in behavioural measures.
c. The tendency of humans to perceive the world as part of their actions (“with a hammer in his hand, a man only sees nails”).
d. A facilitation/inhibition effect of sentences denoting a given body movement direction on the speed of a body movement in the same/opposite direction.

Answer 2

d. A facilitation/inhibition effect of sentences denoting a given body movement direction on the speed of a body movement in the same/opposite direction

Question 3

What is special about the human language brain?
a. There is a particularly strong connection between left inferior-frontal and temporal areas.

b. Neurons in Broca’s region are especially large.
c. The human cortex includes many more pyramidal neurons than the cortex of any other animal.
d. Broca’s region has increased most substantially in the evolution from macaque to human.

Answer 3

a. There is a particularly strong connection between left inferior-frontal and temporal areas.

Question 4

What is semantic grounding?

a. The process of establishing the relationship between the form of a given symbol and objects and actions it can be used to speak about.
b. The process of establishing the relationship between the form of a given symbol and the mental images and concepts it relates to.
c. The process of establishing the relationship between the form of a given symbol and other symbols with which it frequently occurs.
d. The process of establishing the relationship between the form of a given symbol and its abstract semantic features.

Answer 4

a. The process of establishing the relationship between the form of a given symbol and objects and actions it can be used to speak about.

Question 5

The semantic prediction potential …

a. Precedes sentence onset by up to 200 ms.
b. Indexes semantic differences with short latencies (ca 200 ms) upon hearing critical words.
c. Indexes semantic differences between expected words.
d. Precedes meaningful actions if they involve the use of tools.

Answer 5

c. Indexes semantic differences between expected words.

Question 6

What is an ‘auto-associative’ network?

a. A deep neural network with random connectivity and associative learning, which adjusts the number of its layers automatically.
b. A network with connections between its neuron members, even within one network part or ‘layer’.
c. A network that associates neurons with each other based on a computerized and therefore automatic algorithm.
d. A network that realizes Hebbian co-occurrence learning in addition to weakening of connections after uncorrelated activation.

Answer 6

b. A network with connections between its neuron members, even within one network part or ‘layer’.

Question 7

What is meant by ‘fast mapping’?

a. The ability of Hebbian learning to account for language acquisition in early infancy.
b. The rapid learning of mental maps in macaques and great apes.
c. The surprising ability of deep networks with error backpropagation to account for fast learning if neurons with radial threshold functions are used.
d. The human specific rapid learning of novel words after just a few or even just one presentation(s).

Answer 7

d. The human specific rapid learning of novel words after just a few or even just one presentation(s).

Question 8

What is a deep neural network?

a. A distributed auto-associative network with localist representations.
b. A network which can be used for simulating deep thought and language.
c. A network that can add more layers if the learning task requires this.
d. A hetero-associative network with several layers, typically 6 or more.

Answer 8

d. A hetero-associative network with several layers, typically 6 or more.

Question 9

Neural networks are sometimes seen as not biologically realistic enough. Why?

a. Because few neural networks model connectivity between areas and between local neurons.
b. Because the brain is so complex that it can never be approximated by a computer simulation.
c. Because they do not include real neurons, but only artificial ones.
d. Neural networks are called ‘neural’ because they are sufficiently realistic biologically.

Answer 9

a. Because few neural networks model connectivity between areas and between local neurons.

Question 10

Which of the following is not a constraint for making neural networks more biologically plausible?

a. Simulating the local interplay between excitatory and inhibitory neurons within one cubic millimeter of cortex.
b. Realizing long term depression in addition to long term potentiation.
c. Implementing between-area connectivity as revealed by tractography.
d. Increasing the depth of the network.

Answer 10

d. Increasing the depth of the network.

Question 11

Behavioral experiments indicate that, after a person has moved many beans from a tray in front of her/him to one placed slightly further way, the following happens:

a. Whole sentences can be more easily understood, but not single words.
b. The sentence “She gives all her money away” is processed more slowly than the sentence “She steals some money from her friend”.
c. Language understanding is speeded substantially, because the motor activation facilitates language processing.
d. The sentence “She opens the drawer” is processed more quickly than the sentence “She closes her drawer”.

Answer 11

b. The sentence “She gives all her money away” is processed more slowly than the sentence “She steals some money from her friend”.

Question 12

Which features of 3-layer perceptrons or ‘parallel distributed processing’ models should be altered in order to increase the biological realism of these networks?

a. All of the features mentioned in the other answers
b. Error backpropagation learning
c. Restriction to 3 layers or compartments

d. Hetero-associative links between layers (areas) and no links within layers
a. All of the features mentioned in the other answers b. Error backpropagation learning

Answer 12

a. All of the features mentioned in the other answers

Question 13

What are features of ‘whole brain modelling’?

a. The backpropagation learning rule is typically used.
b. Whole brain models are frequently used to model the outcome of perceptual learning experiments with fMRI.
c. The ‘neurons’ of the model simulate entire brain parts or areas.
d. The inhibitory links between local neuron populations are implemented to control the level of activity at a global level.

Answer 13

c. The ‘neurons’ of the model simulate entire brain parts or areas.

Question 14

Which of the following methods is NOT an established method of aphasia therapy?

a. Neurolinguistic programming or NLP.
b. Naming and word picture matching.
c. Constraint induced aphasia therapy or CIAT.
d. Melodic intonation therapy or MIT.

Answer 14

a. Neurolinguistic programming or NLP.

Question 15

Brain lesions can have severe effects on language. Lesions in the lateral motor cortex have been reported to …

a. Lead to a deficit in processing tool related nouns.
b. Lead to a general deficit in producing language – but not in understanding language.
c. Lead to a general deficit in producing and understanding language.
d. Lead to a deficit in verb processing.

Answer 15

a. Lead to a deficit in processing tool related nouns.

Question 16

1. Why did Wittgenstein speak about ‘language games’?

A. To bring into prominence the fact that the speaking of a
language is part of an activity, or of a form of life.
B. To emphasise the role of game theory in linguistics.
C. To bring into prominence that language is a leisure activity.
D. To highlight that game like activities are important in
foreign language learning.

Answer 16

A. To bring into prominence the fact that the speaking of a

Question 17

2. What happens to the brain response if a semantically implausible target sentence is put into a context that makes it plausible, e.g. “The crocodile ate an avocado” (target
   sentence) ; “A crocodile once decided to become a vegetarian” (context sentence).

A. The unexpected word (“avocado”) elicits a mismatch negativity.
B. The unexpected word (“avocado”) elicits a typical N400
C. The unexpected word (“avocado”) elicits an even larger N400 than normally (without any context sentence).
D. The unexpected word (“avocado”) elicits a reduced or even no N400.

Answer 17

D. The unexpected word (“avocado”) elicits a reduced or even no N400.

Question 18

3. Which of the following components is NOT covered by standard speech act descriptions?
   A. The propositional content,
   B. The speech act type or illucutionary role,
   C. The linguistic competence of the communication partners,
   D. The utterance.

Answer 18

C. The linguistic competence of the communication partners,

Question 19

4. “This is the end of the exam”. Which statement best describes the communicative function of this sentence?

A. It is used to terminate the exam.
B. It is used to label a section of a video recording.
C. Dependent on communicative context, it can be used to perform different speech acts.
D. Dependent on communicative context, it can be used to perform different declarative speech acts.

Answer 19

C. Dependent on communicative context, it can be used to perform different speech acts.

Question 20

What are key differences between the speech acts of naming and requesting?

A. As the same words can be used by speakers to name and request,
there are no relevant differences.

B. Pragmatic functions are largely unimpaired in aphasia; therefore,
any possible difference between speech acts are not relevant.

C. When naming an object, there are no specific expectations towards
the partner, whereas requesting is characterized by the expectation
of being given the requested entity.

D. The sequence structures of these speech acts differ in complex
ways..

Answer 20

C. When naming an object, there are no specific expectations towards
the partner, whereas requesting is characterized by the expectation
of being given the requested entity.

Question 21

1 What is aphasia?
A. A brain disease affecting language in different people of all ages
B. A language deficit caused by focal brain lesion after completion of language acquisition
C. A subtype of dementia
D. A specific syntactic deficit of neurological origin with lexicon and pragmatic functions being spared

Answer 21

B. A language deficit caused by focal brain lesion after completion of language acquisition

Question 22

2 Which of the following methods is NOT a standard in current aphasia therapy?
A. Naming and word picture matching
B. Government and binding therapy
C. Melodic intonation therapy
D.Linguistic tasks such as sentence completion and semantic classification

Answer 22

B. Government and binding therapy

Question 23

3 Which of the following rules for aphasia therapy is NOT motivated by neuroscience research?
A.Intensity principle: practice as much as possible in as short a time as possible
B. Learned nonuse prevention: guide patients to take advantage of their (linguistic, motor) capabilities
C. High alertness constraint: make sure patients are maximally attentive at all times
D. Use language in pragmatic action contexts

Answer 23

C. High alertness constraint: make sure patients are maximally attentive at all times

Question 24

4. What constitutes strong evidence for the effectiveness of a specific speech
   language therapy method?

A. If patients perform better after therapy as compared to before on standardised
clinical tests and on tests of pragmatic communicative linguistic function.
B. If neurophysiological and neurometabolic responses to linguistic stimuli
increase when applying the specific method of speech language therapy.
C. If a randomised controlled trial shows that application of the method is
accompanied by relevant performance increase, but a control therapy is not, or
significantly less.
D. If the relatives of the patient and at least three independent MDs agree that
there was strong improvement across therapy, and if this result is confirmed by
clinical linguistic testing.

Answer 24

C. If a randomised controlled trial shows that application of the method is
accompanied by relevant performance increase, but a control therapy is not, or
significantly less.

Question 25

5. Is behavioural treatment of aphasia effective in improving speech
   language deficits?

A. Yes, aphasia therapy works, but only at the acute stage. In chronic aphasia more than one year after disease onset, any effects of therapy are minimal or absent.
B. No, aphasia therapy does not work, because neurological tissue cannot be restored by behavioural exercises. Neurostimulation
methods (TMS, tDCS) are necessary.
C. Yes, some intensive methods of aphasia therapy were shown to be
effective even at the chronic stage.
D.Large randomized controlled trials performed so far could not prove
the efficacy of behavioral treatment of aphasia.

Answer 25

C. Yes, some intensive methods of aphasia therapy were shown to be
effective even at the chronic stage.

Question 26

6. Which is the most crucial factor for the success of Constraint Induced Aphasia Therapy (or Intensive Language Action Therapy)?
   A.The number of constraints used during therapy.
   B. The embedding of linguistic utterances into social communication.
   C. The warm and friendly atmosphere created in conversation.
   D.This issue is not yet fully clear and needs further investigation.

Answer 26

B. The embedding of linguistic utterances into social communication.

Question 27

1. Which of the models named below is not a model of speech production?
A. Levelt’s model
B. Dell’s model
C. Frazier’s garden-path model
D. WEAVER++

Answer 27

C. Frazier’s garden-path model

Question 28

2. What structural phenomenon is exemplified by the following example sentence: “The man that the girl that the cat fed greeted laughed.”
A. center embedding
B. reduced relative clause ambiguity
C. lexical ambiguity
D. word category ambiguity

Answer 28

A. center embedding

Question 29

3. What ambiguity is not a structural ambiguity?
A. word order ambiguity
B. referential ambiguity
C. attachment ambiguity
D. thematic role ambiguity

Answer 29

B. referential ambiguity

Question 30

4. What is not a key stage in the Levelt model of language production?
A. lexical selection
B. phonetic encoding
C. conceptual preparation
D. integration with discourse

Answer 30

D. integration with discourse

Question 31

5. What eye-gaze pattern did Trueswell et al. (Cognition, 1999) observe in children in response to visual-context disambiguation (1 vs. 2 referents) for sentences (c1): Put the frog on the napkin in the box compared with (c2) Put the frog that’s on the napkin in the box?

A. More looks to the incorrect goal for sentences c1 than c2 and slightly more looks to that goal for 2-referent than 1-referent contexts

B. Less looks to the incorrect goal for sentences c1 than c2 and slightly more looks to that goal for 2-referent than 1-referent contexts

C. More looks to the incorrect goal for sentences c2 than c1 and slightly less looks to that goal for 2-referent than 1-referent contexts

D. Identical gaze patterns

Answer 31

A. More looks to the incorrect goal for sentences c1 than c2 and slightly more looks to that goal for 2-referent than 1-referent contexts

Question 32

6. A central postulate of modularity of the mind is what?
   A. that cognition is embodied
   B. that cognitive processes are informationally encapsulated
   C. that the language and visual systems can directly interact with each other
   D. that language processing is strongly situated

Answer 32

B. that cognitive processes are informationally encapsulated

Question 33

1) What characterized the A-system of cortico-cortical connections between pyramidal neurons
a) Long axons – basal dendrites
b) Local axons – basal dendrites
c) Long axons – apical dendrites
d) Local axons – apical dendrites

Answer 33

c) Long axons – apical dendrites

Question 34

2) Which part of the brain is not part of the perisylvian language areas?
(a) Articulatory motor cortex
(b) Visual cortex
(c) Wernicke’s area
(d) Auditory cortex

Answer 34

(b) Visual cortex

Question 35

3) Which statement about the connection between the processing of abstract and concrete words is NOT correct?

a) Both (concrete and abstract words) are learnt when used to speak about events/objects/actions/etc.
b) The variability of the sensorimotor patterns that foster semantic grounding is typically high for concrete and low for abstract symbols.
c) For abstract meaning there are family resemblance pattern and the
links to the symbol are weak while in concrete meaning semantic
feature overlap and strongly link to the symbol.
d) The differences in correlation structure of abstract vs. concrete words lead to different neuronal & cognitive mechanisms in the processing of these words.

Answer 35

b) The variability of the sensorimotor patterns that foster semantic grounding is typically high for concrete and low for abstract symbols.

Question 36

4) In the study with tumor patients by Dreyer, Picht et al 2020, category-specific semantic deficits were found for action words. Which objection of the “grounding debate” can be weakened or even invalidated with this?
a) Semantic grounding effects in brain & behavior are not
reproducible
b) Semantic grounding effects in brain & behavior are
epiphenomenal
c) Semantic grounding effects in brain & behavior are not functional
or causal
d) Semantic grounding effects in brain & behavior are not flexible

Answer 36

c) Semantic grounding effects in brain & behavior are not functional
or causal

Question 37

5) Different speech acts…
a) Do not have specific brain signatures when
conveyed by the same linguistic form.
b) Are associated with different signatures but
only when conveyed by gestures.
c) Are associated with different signatures which reflect differences in their action sequence trees.
d) Have specific brain signatures that become
evident only when they are predicted by context.

Answer 37

c) Are associated with different signatures which reflect differences in their action sequence trees.

Question 38

6)	Which	Maxim	of	Grice	does	the	following	sentence	violate:	
‘Humans	can	obviously	fly’?
a) Quality
b) Quantity
c) Relation
d) Manner

Answer 38

a) Quality

Question 39

7) Which of these is NOT a typical method in standard behavioral speech language therapy:
a) Matching sentences to pictures
b) Finding synonyms
c) Requesting objects
d) Naming and describing pictures

Answer 39

c) Requesting objects

Question 40

8)	The	greater	efficacy	of	ILAT/CIAT	therapy	compared	to	conventional	
aphasia	therapy	lies	essentially	in	the:
a) Embedding	in	language	games
b) Extensive	use	of	prosodic	cues
c) Greater	frequency
d) Addressing	the	learned	non-use

Answer 40

a) Embedding in language games

Question 41

9) Which is NOT a limitation of whole brain models?
a) They are limited to area level (as lowest level)
b) They don’t capture representations
c) They don’t learn tasks
d) They use backpropagation

Answer 41

d) They use backpropagation

Question 42

10)	Which	is	NOT	true	of	the	brain-constrained	cortex	
model	covered	in	the	podcasts?	
a)	It	is	auto-associative
b)	It	uses	backpropagation
c)	Some	versions	have	used	spiking	cells
d)	It	is	sparse

Answer 42

b) It uses backpropagation

Question 43

11) Which of the ‘neurons’ or functional units is the most realistic one:
a) Mean Field Neuron
b) McCulloch-Pitts Neuron
c) Node in a Whole Brain Model
d) Spiking Neuron

Answer 43

d) Spiking Neuron

Question 44

12) What happens if you move from a 3-layer network
to a deep neural network?
a. It gets more biologically realistic
b. It becomes auto-associative
c. It becomes hetero-associative
d. It gets capable of performing better in complex tasks

Answer 44

d. It gets capable of performing better in complex tasks

Question 45

13) Which of the following is not a type of local ambiguity?
a) The horse ran past the barn fell
b) He put the shoes by the door in his bag
c) The woman played with the baby in the green shirt
d) Die Prinzessin grüßt den Mann (The princess(accusative/object) greets the man(nominative/subject)

Answer 45

c) The woman played with the baby in the green shirt

Question 46

14) What structural phenomenon is exemplified by the following example sentence: “The man
that the girl that the cat fed greeted laughed.”

a) center embedding
b) reduced relative clause ambiguity
c) lexical ambiguity
d) word category ambiguity

Answer 46

a) center embedding

Question 47

15) Which behavioral measure is NOT an offline measure?
a) eye tracking
b) accuracy questionnaires
c) plausibility ratings
d) neuro-psychological tests

Answer 47

a) eye tracking

Question 48

16) Which is NOT a measurement that we can compute from eye movement records?
a) log-gaze probability ratio
b) fovea size
c) first fixation duration
d) AOI inspection

Answer 48

b) fovea size

Question 49

17) According to Frazier’s Garden Path Theory, which of the following principles does not guide syntactic processing?
a) When processing sentences we construct a single interpretation at a time.
b) Interpretations of sentences are re-analysed on the basis of syntactic conflict.
c) Interpretations of sentences are reanalysed on the basis of low plausibility.
d) World knowledge and visual scene immediately influence syntactic processing

Answer 49

d) World knowledge and visual scene immediately influence syntactic processing

Question 50

18) Tanenhaus et al investigated the influence of visual scene information in syntactic processing using structurally ambiguous sentences such as ‘Put the apple on the towel in the
box’. Which of the following statements is false?

a. Participants used visual scene information to help resolve structural ambiguity.
b. Visual scene information rapidly influenced syntactic processing.

c. Results showing the influence of visual scene information on syntactic processing support the
hypothesis that syntactic processing is modular.

d.The	auditory	presentation	of	stimuli	allowed	Tanenhaus	et	al	to	investigate	influence	of	visual	
context	information	(as	opposed	to	semantic	information)	on	syntactic	processing.

Answer 50

c. Results showing the influence of visual scene information on syntactic processing support the
hypothesis that syntactic processing is modular.

Question 51

19) Which of the following statements about the N400 component is true?
a) N400 is traditionally seen as an index of syntactic violations.
b) The N400 component is only elicited in linguistic contexts.
c) N400 is a graded index of semantic mismatch.
d) Semantically strange sentences always elicit an N400 response.

Answer 51

c) N400 is a graded index of semantic mismatch.

Question 52

20) Which statement about the integration of visual information on semantic processing is false?

a) We prefer to use world-knowledge of stereotypical thematic roles over depicted information.
b) Motor behaviour (eye-gaze) is modulated by the semantic interpretation of motion words.
c) Even subtle relations between spatial distance and semantic similarity impact real-time
semantic interpretation of abstract sentences.
d) Semantic interpretation is rapid and incremental even in rich visual contexts.

Answer 52

a) We prefer to use world-knowledge of stereotypical thematic roles over depicted information.

Question 53

21) What is a characteristic of lexical processing?
a) We wait until word completion before recognition
b) We compare competing representations in serial fashion
c) Candidate words are not isolated from each other’s activation
d) Acoustic-phonetic information doesn’t modulate lexical processing

Answer 53

c)Candidate words are not isolated from each other’s activation

Question 54

21) What is a characteristic of lexical processing?
a) We wait until word completion before recognition
b) We compare competing representations in serial fashion
c) Candidate words are not isolated from each other’s activation
d) Acoustic-phonetic information doesn’t modulate lexical processing

Answer 54

c)Candidate words are not isolated from each other’s activation

Question 55

22) Van Berkum et al. (2008) study demonstrated that:

a) … a comprehender’s level of education influences their real-time comprehension and visual
attention.
b) … social cues behave similar to lexical semantic cues in linguistic contexts.
c) … intra-linguistic semantic information and extra-linguistic world knowledge are integrated
within the same time frame during sentence interpretation.
d) … semantic interpretation is separate from nonlinguistic elements of meaning

Answer 55

b) … social cues behave similar to lexical semantic cues in linguistic contexts.

Question 56

23) Which of the following is not a property of the Socially Co-ordinated Interplay Account?
a) Properties of the comprehender
b) Properties of the speaker
c) Visual scene
d) Theory of Mind

Answer 56

d) Theory of Mind