Student Vt 2009
Examensarbete, 15 hp
The Search for an Elusive Word:
Verbal Search Procedures in Tip-of-the-Tongue States
Abstract
The tip-of-the-tongue (TOT) state refers to the familiar predicament of being unable to retrieve a specific word at a required moment, while being certain of knowing the word. TOT states may be caused by incomplete activation levels of an elusive target word, or by the initial retrieval of a phonologically related word blocking retrieval of a target word. In this thesis, a spoken response experiment designed to monitor the search procedure was used to elicit and examine the TOT state. The results showed that related words were absent in most TOTs, and that partial phonological information was retrieved initially in a majority of TOTs. These results support the
partial activation hypothesis’ notion that incomplete activation levels of the target
Acknowledgements
First and foremost, I would like to express my gratitude to my supervisor Leila Kantola for providing me with many valuable comments and suggestions on how to improve this thesis. I have learnt a lot from you during this process. Thank you, Leila. I would also like to thank Fredrik Karlsson at the Department of Language Studies for helping me with arrangements concerning the sound studio.
The language production system is very complex, as production probably comprises several stages and processes. An influential language production model by Levelt (1989) proposes that the processes of speech production are incremental, and constitute three wide areas named conceptualization, formulation and encoding. That is, we first decide what we want to say, we then formulate the concept mentally, and finally, we encode the lexical form into sounds so that we can articulate it. Because language production is a rapid process (Levelt, Roelofs, & Meyer, 1999), it is difficult to isolate the processes from each other, which in turn makes research on the production processes complicated. There is however a window into language production that can provide an insight to its underlying complexity, and that is the study of speech errors. Even though speech in general tends to be fast and accurate, dysfluencies and errors are not uncommon (Garnham, Shillcock, Brown, Mill, & Cutler, 1981), and speech errors span across different linguistic levels. Errors can occur on a single phoneme or phonemic clusters, as well as on morphemes, syllables, words or even phrases. There are also various types of errors, such as word substitution errors, hesitations, blends, false starts and repetitions. Regularities found in error patterns have been regarded as evidence for different language production processes (Butterworth, 1982; Edeltrud, 1999; Harley & MacAndrew, 1995; Harley & MacAndrew, 2001). By isolating and revealing finer levels of speech production, errors thus shed light on how the production system works.
One particular type of speech error that may help establish which stages are involved in the process of retrieving a word from the mental lexicon, the lexical retrieval process, is a phenomenon named ‘tip-of-the-tongue’ (TOT). A frequently quoted characterization of a TOT state, is given by William James (1890, p.251):
“Suppose we try to recall a forgotten name. The state of our consciousness is peculiar. There is a gap therein; but no mere gap. It is a gap that is intensely active. A sort of wraith of the name is in it, beckoning us in a given direction, making us at moments tingle with the sense of our closeness, and then letting us sink back without the longed-for term.”
TOT demonstrates a breakdown in lexical retrieval, which is one of the processes involved in formulation. A typical TOT state may be described as an occasion when a speaker knows that a particular word for example refers to “scents that are transmitted between individuals of the same species and carry a certain message” (i.e., the speaker is aware of the semantic representation) but is still unable to retrieve the target word feromoner (the phonological form of the target). The speaker may however be able to retrieve certain information about the target word itself, for example, the first phoneme /f/ of the word or retrieve another word similar in sound, for example freoner, instead of the right word. To be precise, TOT refers to the temporary inability to retrieve the phonological form of a word that a speaker feels absolutely certain exists in the mental lexicon, and this inability is accompanied by a sense of imminent retrieval (R. Brown & McNeill, 1966). The TOT phenomenon is something speakers tend to experience relatively frequently in conversations – in particular when the target is a low-frequency word (Harley & Bown, 1998); TOTs seem to occur on average once a week (Burke, MacKay, Worthley, & Wade, 1991; Reason & Lucas, 1984) and have also been suggested to be universal (Brennen, Vikan & Dybdahl, 2007; Schwartz, 1999, 2002).
often be remembered (R. Brown & McNeill, 1966; Koriat & Lieblich, 1974; Rubin, 1975). TOTs seem to attest that the phonological representation of a word in our mental lexicon may not be an indivisible entity, but is instead built up from smaller parts such as phonemes and syllables. The partial information retrieved during TOTs also shows that certain units or characteristics of a word are more prominent in our mental lexicon than others. Furthermore, during TOT states, semantically and phonologically related words – i.e., words that are similar in meaning and in sound, respectively – may also be retrieved, although the desired word cannot be retrieved (R. Brown & McNeill, 1966; Burke et al., 1991; Cohen & Faulkner, 1986).
Lexical retrieval can be studied at a slower pace when examining TOTs since a complete retrieval of the intended word does not usually occur immediately when a speaker is in a TOT state. It is almost as if the retrieval process has been frozen in TOT; the speaker has accessed semantic information about the target word, and perhaps partial phonological information, but the partial information is not sufficient for the full target word to be accessed. There is therefore an emerging consensus among researchers that TOTs are a partial failure of lexical retrieval (e.g., Astell & Harley, 1996; Harley & Bown, 1998; James & Abrams, 2002). Harley and Bown describes that the majority of speech production models seem to recognize a non-phonological specification that is unique for individual words, and that mediates between the semantic and phonological representations. Specifically, when producing a word, the speaker first activates the appropriate semantic information. Next, in an intermediate level between the semantic and phonological representation, words are represented as non-phonological lexical units – lemmas. These are selected from the semantic specification and convey the desired concept. Results indicate that TOTs occur because of an inability to accomplish the second stage of word retrieval (Astell & Harley, 1996; Harley & Bown, 1998), i.e., an inability to retrieve the phonological form – the lexeme – that corresponds to the lemma. It appears as if there is successful activation of the target lemma but not of the lexeme. In sum, the TOT phenomenon illustrates how we can know the meaning of a word but at the same time not recall the word itself. The retrieved partial information seem to reflect how phonological forms are represented in the mental lexicon, and how the search procedure may operate, as well as that the breakdown often is a partial failure at a later stage of lexical retrieval. Hence, the TOT state allows research on nuances of the stages involved in lexical retrieval.
The study conducted by R. Brown and McNeill (1966) was the first to introduce experimental methodology in the study of TOTs. Here TOT states were induced by presenting participants with definitions of low-frequency words and instructing participants to, when in a TOT state, complete a response sheet by providing specific information about the target words. Their results were similar with findings from observational studies (Woodworth, 1934) on naturally occurring TOTs. Participants showed a greater-than-chance ability to provide partial information about the target. The distinction between phonological and semantic access was first introduced in this study. It was also found that the words produced instead of the target in TOT states were predominantly phonologically related to target words, although participants also produced semantically related words.
word, referred to as a blocker, is the cause of TOT states. The theory of blocking applies specifically to words that are phonologically related to the target, although words with both a semantic and phonological similarity may hinder target retrieval within this view (A. Brown, 1991). The implication of this theory is that target word retrieval fails because the blocker is competing and interfering with the retrieval of the target. This related word is more accessible than the target because it is used more frequently and may thereby become activated faster than the target. Once activated, it blocks target retrieval; even though the speaker is aware that the related word is not the target, the elevated activation blocks the access of the target and thus causes a TOT state (Jones, 1989; Jones & Langford, 1987). According to the blocking hypothesis, the activation of the phonologically related word can also inhibit or delay TOT resolution (A. Brown, 1979). In the study by A. Brown, a phonologically related word was presented to the participants before they were given the definition of the target word. The presentation of the related word led to a delay in target retrieval, as well as a decrease in the probability of target retrieval, compared to a condition where an unrelated word was presented. However, these target retrieval difficulties were only found when participants expected that the presented word could be the target; that is, the effect disappeared when participants were aware that the presented word could not be the target (Roediger, Neely, & Braxton, 1983). There have been other studies (Jones, 1989; Jones & Langford, 1987) where TOTs are elicited by the presentation of phonologically related cues in conjunction with the presentation of target definitions (both prior and post definition presentation) that support blocking as an explanation for TOT. It was found that participants tended to experience TOTs more frequently when phonologically related words were used as cues compared to when the cues were not phonologically related to the target.
shown their blocker on a screen. This did however not decrease retrieval rates. This also contradicts the blocking explanation for TOT since the blocker should obstruct retrieval. Further, results from a study by Burke et al. (1991) with participants from different age groups, indicate that the frequency of experienced phonologically related words during TOTs decrease with age, but the frequency of TOT states on the other hand seems to increase or remain constant with age. This was considered a complication for the blocking hypothesis by Burke et al. since if phonologically related words are the actual cause of TOT states, phonologically related words should occur at the same frequency as the TOTs.
Both the blocking hypothesis and the partial activation hypothesis state that, although more than one word may be activated, the retrieved word is the word that has received the highest level of activation. However, the blocking hypothesis explains the occurrence of TOTs with a competitor that receives more activation and inhibits the intended target, whereas the partial activation hypothesis assumes that inadequate activation of the target leads to TOTs. The transmission deficit model (Burke, MacKay & James, 2000; Burke et al., 1991) is a model that similarly to the partial activation hypothesis proposes that it is an inadequate activation level of the target that induces TOTs – and not a competitor blocking the target. The retrieval difficulty is caused by weak connections between the semantic and phonological systems within this view (Burke et al., 1991). The details of this theory will not however be discussed here.
The purpose of this study is to examine whether incomplete activation of a target can be seen to cause TOT states. Congruent with the assumptions of the partial activation hypothesis, it is hypothesised that TOTs originate when the target receives incomplete levels of activation and that TOTs are not caused by a related word blocking target retrieval. According to the partial activation hypothesis, TOTs frequently occur without any related words being retrieved. If the participants in the study are not found to produce related words initially when in TOT states, TOTs are not likely caused by related words blocking the target retrieval. If, on the other hand, the participants are found to produce words that are related to the target words initially when in TOT states, this finding does not support the partial activation account of the cause of TOTs but gives support to the blocking hypothesis.
Experiment
In TOT induction studies, when a participant cannot produce a target word corresponding to a given definition, it might be that this happens because the word is unknown to the participant. In this study, to lower the possibility of data loss due to participants never having been exposed to some words, the experiment consisted of a presentation phase and a testing phase. At the presentation, participants were asked to listen to a pre-recording of 30 words with their corresponding definitions. Three days after, in the testing phase, participants were given the same definitions and asked to provide the corresponding word.
A pilot test with four participants confirmed that is was possible to use the method of spoken responses, and the two-part approach. It did however lead to an adjustment of the experiment. At the pilot, participants had received instructions to not only say what they knew about the word, but they also learned that speakers in TOT states might retrieve phonological and semantic neighbours instead of a given target. This resulted in participants consciously searching for words to fit a given definition, or trying to find synonyms to one of the words in a given definition. This part of the instructions was therefore excluded from the experiment. Hence, contrary to prior studies (e.g., R. Brown & McNeill), participants in this study did not receive any specific questions or instructions other than to report what they knew of the irretrievable target word.
Method
Participants
25 students at Umeå University were recruited for the experiment. All of the participants were native speakers of Swedish with a mean age of 25.7 (Female: N=13, mean age=24.2, age range=22-28; Male: N=5, mean age=29.8, age range=27-32). The participation was voluntary and participants were not compensated for their time. There were seven exclusions in total. In four of the cases, this was due to participants taking part in the presentation phase yet failing to attend the actual testing occasion. In three cases, it was due to failure of the equipment in the sound studio on the test occasion. In total, 18 participants completed the experiment.
Materials and design
The materials included 30 Swedish words and their corresponding definitions. The words consisted of 23 nouns, six verbs and one adjective, with an average syllable length of 3.45 (range 2-5). Most of the target words were low-frequency words (according to Allén (1971), see Appendix). To increase the likelihood of correct answers in order to maintain participant motivation during the second part of the test, some target words were of higher frequency. It was assessed that when feeling confident, participants would be more likely to “think aloud” during the test. The definitions were generated from Nationalencyklopedins ordbok, with 13 of the definitions adapted to make them as clear as possible.
used for the presentation phase and the other used for the actual test.
Test materials for the presentation phase. For the presentation phase of the
experiment, a list with the 30 words and their respective definitions was compiled and recorded. The recorded material was edited in the audio editing software Audacity to create a two second pause between each unit (i.e., word + definition), and a 0.65 second pause between each target word and its definition.
To prevent participants from repeating the words to themselves after the session, which could decrease the probability of TOTs, intervention tasks were created for the presentation phase. The intervention materials included 25 common nouns and verbs (M of syllable length: 2.3). The material also included 10 math exercises that varied in degree of difficulty.
Stimulus materials. An audio file containing the recorded definitions from the
presentation phase was created for the TOT eliciting test. The definitions were rearranged. The first three definitions corresponded to words of higher frequency in an attempt to increase the likelihood of some correct responses in the beginning of the test phase, in order to increase participant confidence. The list was edited in Audacity so that there would be a 30 second long pause in between every definition, giving participants time to respond.
Procedure
Presentation phase. The experiment design of the presentation phase comprised two
stages. First, one at a time, each participant was presented with the pre-recorded material via headphones. After the presentation was completed, intervention tasks immediately followed. Participants were assigned a memory task where the objective was to memorize words from a list of 25 words, with two minutes to their disposal. This was followed by an interval of one minute during which the participants were given a sheet of paper consisting of 10 math exercises. Participants received instructions to solve as many as possible, within a time frame of two minutes. After two minutes of solving math problems, participants received instructions to turn the sheet over and write down as many words as possible from the list previously read, and were given one minute to do so. After completion of the tasks, an appointment for the next part of the experiment was set up.
Test phase. The TOT-test consisted of a single task and took place three days
after the presentation session. Participants were informed that they via headphones would be presented with the definitions they had listened to at the presentation phase, and that the end of each definition would serve as a cue for each response. They were also informed that each definition would be followed by a 30 second long pause, during which they had a chance to provide their answer. Participants were instructed to say the word corresponding to the definition aloud as soon as they knew the word. If they could not produce the target word, but felt as if they did know the word and were close to producing it, they were encouraged to say aloud everything they knew about the word as quickly as possible. The instructions emphasized the importance of “thinking aloud” at all times during this test. After the instructions had been presented, participants were given the opportunity to ask questions.
sound-treated recording room in the studio. The experimenter was present in the studio during the test phase, seated in the adjoined control room on the opposite side of the participant, with a window dividing. To ensure that participants tried to “think aloud” at all times, the experimenter held up a sign through the dividing window saying just that, occasionally during the test.
After completion of the test, participants were debriefed in cases when any obscurities had appeared (see criteria below). The entire procedure lasted approximately 25 to 30 minutes per participant.
TOT inclusion criteria
To qualify as a TOT, it was required that participants did not appear to be consciously searching for synonyms to words in a given definition, or search for a word to fit the description. In addition, one of the three following criteria had to be fulfilled:
(i) A participant stated that he or she knew the word and felt very close to
producing it.
(ii) An answer included partial phonological information about the target
word.
(iii) An answer included a word, a real word or a non-word, that was
phonologically or semantically related to the target word.
In cases when participants did not know the target word and the answer included partial phonological information about the target, i.e., the answer fulfilled criteria (ii) above, though participants expressed some hesitation (e.g. by making an utterance of the type “But I’m not sure”), a post-test questioning was administered verbally. This was done in order to eliminate any ambiguity as to whether the person was in a TOT state or not. Only responses where the participant stated that the word was close to being retrieved, but he or she just could not access the word, were included in the analysis.
Results
Altogether 433 answers (i.e., not blank responses) were recorded on 540 trials (80.2%). 32 of these answers (7.4%) were analysed as TOTs, which corresponds to an average of 1.8 TOTs per participant (range 0 – 6), see Table 1.
The responses were classified into four categories: correct targets – the production of the word defined; incorrect targets – the production of a word not defined and without phonological or semantic resemblance to the word defined; blank
responses – instances when no responses or “I don’t know”-responses were provided;
and the fourth category refers to TOT states (see criteria above).
Response category Number of correct Number of incorrect Number of blank Number of T O T Number of Target index ta rgets ta rgets repsonses responses resolved T O Ts Abdikera 13 2 2 1 0 Akademiker 12 3 1 0 -Amalgam 15 2 1 1 1 Ambivalens 8 5 4 1 0 Appellera 9 0 6 3 3 Arbitrage 0 11 5 2 0 Blasfemi 1 5 9 3 0 Celibat 14 2 1 1 0 Chassi 12 3 3 0 -Deklaration 16 1 0 1 1 Extrovert 9 6 2 1 0 Facit 13 1 4 0 -Feromoner 9 4 2 4 2 Friktion 17 0 1 0 -Hallucination 14 2 1 1 1 Implikation 2 8 8 0 -Inkompetens 3 9 6 0 -Kannibal 18 0 0 0 -Kapsejsa 8 4 4 2 2 Kondoleans 12 2 3 1 1 Likvidation 6 5 7 0 -Obduktion 17 0 1 0 -Pacifism 9 3 5 1 1 Perforering 11 3 2 2 1 Postulera 1 11 4 2 0 Regression 10 2 4 2 0 Rekognosera 6 7 4 1 1 Reprimand 7 0 10 1 0 Sta ffli 16 0 1 1 1 V idimera 7 5 6 0 -295 106 107 32 15 Table 1:
Distribution of Responses and Resolved T
OT
all TOTs.
For the purpose of this study, the TOT responses were separated into three different types: answers containing related words, answers containing partial phonological information and answers were no target information was provided. TOT states primarily occurred without the presence of related words; 68.8% of TOTs were not accompanied by an initial retrieval of a related word. 31.3% of TOTs occurred in conjunction with the initial retrieval of a related word, and 59.4% occurred with the initial retrieval of partial phonological information. TOTs without any partial information or related words present, occurred in 9.4% of the instances (see Figure 1).
TOT states starting with the production of related words are presented in Table 2. Each row in the table corresponds to an individual search procedure.
The results obtained show that the related words retrieved during TOT states were predominantly phonologically related. As a minimum criterion in order to be defined as a phonologically related word, a word had to share the same initial phoneme, or the same final phoneme cluster as the target (see Harley, 1984; Harley & Bown, 1998 for similar definitions). Phonologically related words were produced in seven instances (21.9% of all TOTs); semantically related words were produced in one instance (3.1% of all TOTs); and words phonologically and semantically related were produced in two instances (6.3% of all TOTs). To illustrate the different word categories, a phonologically related word to feromoner was for example freoner. An example of a semantically related word was motion for appellera; and a semantically and phonologically related word was sondera for rekognosera.
Figure 1: Type of information retrieved during TOT states.
For the current purposes of testing whether TOT states are caused by related words, and given that the blocking hypothesis applies particularly to words that are phonologically related, words that are both semantically and phonologically related will fall under the category phonologically related.
31.3% 59.4% 9.4% 0% 10% 20% 30% 40% 50% 60% 70%
Related words Partial information No information
T
OT
s
A full target retrieval, i.e., TOT resolution, as a result of the production of phonologically related words (words both phonologically and semantically related are also included) occurred in four instances (44.4% of all TOT responses starting with the production of phonologically related words), whereof three (33.3%) were immediate retrievals of the complete target. A TOT resolution stands for the production of the correct target after being in a TOT state. TOT resolution also includes the production of own-target words: words produced by participants as what they considered the correct pronunciation of the target defined. A full target retrieval as a result of the production of semantically related words occurred in one instance (100% of all TOT responses in conjunction with production of semantically related words).
In 59.4% of the TOTs, partial phonological information was retrieved. The information was nearly exclusively initial and final phonemes. Table 3 displays TOT states with reported target characteristics. The search procedure is illustrated with each trial in a different row.
Initial phonemes/phonemic clusters or final phonemes/phonemic clusters were produced in 19 instances (59.4% of all TOTs). The production of partial target information led to the retrieval of a larger segment of the target, such as initial phoneme cluster, in six instances (31.6% of all TOT responses in conjunction with partial phonological information). A full target retrieval, i.e., a TOT resolution, as a result of the production of partial target information occurred in nine instances (47.4% of all TOT responses in conjunction with production of partial phonological information).
Table 3: Distribution of TOT Responses Occurring in Conjunction with Partial Phonological
Information
Note. PID = participant identification number.
In total, delayed retrieval of the complete target after subsequent definition(s) had been presented, occurred in two cases. In one case, the TOT occurred in conjunction with a phonologically related word, and TOT resolution occurred after the next definition. In the second case, as mentioned in the previous paragraph, TOT resolution occurred after two subsequent definitions had been presented, and in this case, retrieval was not preceded by any information about the target. Further, delayed resolution of phoneme cluster occurred once after a subsequent definition.
Table 4: Distribution of TOT Responses without Retrieved
Informa-tion Prior to ResoluInforma-tion
Discussion
The obtained results support the partial activation hypothesis. As predicted, the data indicates that TOT states reflect an incomplete activation of a target word. The pattern of findings shows that TOTs predominantly occurred without the production of related words, and that participants instead tended to produce partial phonological information about the target.
According to studies in support of the blocking hypothesis, a word retrieved instead of the target acts as a competitor inhibiting target retrieval, and is as such the cause of the TOT state. Thus, if related words cause TOTs they should reasonably appear in conjunction with TOTs. Since the majority of TOTs in this study occur without semantically or phonologically related words, it is indicated that the retrieval of related words do not block target retrieval and cause TOT states. Other studies (e.g., Heine, Ober, & Shenaut, 1999; Reason & Lucas, 1984) have found similar results, confirming that far from all TOTs appear in conjunction with related words. In this study, phonologically related words occurred in barely one third of all the TOT states (28.1%), and immediate TOT resolution occurred in 33.3% of those cases. Putting this into contrast with the rate of immediate TOT resolution in TOT states where no phonological information was produced, which happened in none of the cases, it seems more likely that the phonologically related words actually facilitated the target retrieval, rather than blocked it. Evidence for the theory that phonologically related words facilitate target retrieval has been found in previous TOT induction studies (e.g., Harley & Bown, 1998; James & Burke, 2000; Meyer & Bock, 1992). The explanation for the presence of related words put forth by the partial activation hypothesis, is that related words in TOT states are merely a consequence of the search procedure when the TOT state is already initiated. That is, it may be that the target-retrieval difficulty itself causes a related word to be retrieved instead of the elusive word. Specifically, the partial activation hypothesis states that related words are retrieved because they receive a higher level of activation than the target.
states accessed partial phonological information while in a TOT. The data demonstrates that speakers in a TOT most of the time can access some parts of the intended word but that the full target cannot be accessed. The data shows that participants could start with the production of initial phonemes and that this often seemed to facilitate the access of a larger segment of the target word. In almost half of these instances, full target retrieval occurred. Related words were not present to hinder target retrieval in these cases. The fact that most TOTs in this study occurred in conjunction with partial phonological information and that a relatively small part of the TOTs were accompanied by related words is in line with the partial activation hypothesis, which suggests that insufficient level of target activation is the cause of TOTs.
By testing the hypothesis using the spoken response format it was possible to get a better insight into the search procedure when participants are trying to access a given target word. Specifically, how the participants conduct the search for a given target, and how retrieval of eventual partial phonological information could lead to retrieval of larger segments, was possible to examine in more detail. It was found that TOT resolution became more likely as the retrieval of phonological segments progressed.
The use of spoken responses has the advantage of being more on-line, and may thus provide more immediate information of the search than when participants are asked to write down the information they come across when trying to remember the target word. A subject of future research could perhaps be a comparison between spoken and written responses in TOT induction studies to distinguish any differences in the obtained information.
Whichever method is used to collect participants’ responses, some information will however undoubtedly be lost. For instance, some participants reported after the test how they for a split second had felt unable to retrieve a target but could remember the first letter or letters. This was instantly followed by a full retrieval. This type of barely noticeable TOT has been labelled micro-TOT (A. Brown, 1991). In this state, TOT resolution is claimed to occur so rapidly that the participant hardly notices the word retrieval difficulty. It does however occur long enough for participants to perceive it. While fascinating, the existence of these micro-TOT states is of course difficult to tap experimentally because participants are not able to verbalise them during the test, as was the case in this study. Even though it is intriguing to be able to record participants’ responses, it is of course impossible to capture every thought.
References
Allén, S. (1971). Frequency Dictionary of Present-Day Swedish 2. Stockholm: Almqvist & Wiksell.
Askari, N. (1999). Priming Effects on Tip-of-the-Tongue States in Farsi-English Bilinguals. Journal of Psycholinguistic Research, 28, (2), 197-212.
Astell, A. J., & Harley, T. A. (1996). Tip-of-the-tongue States and Lexical Access in Dementia. Brain and Language, 54, 196-215.
Brennen, T., Baguley, T., Bright, J., & Bruce, V. (1990). Resolving semantically induced tip-of-the-tongue states for proper nouns. Memory & Cognition, 18, (4), 339-347.
Brennen, T., Vikan, A., & Dybdahl, R. (2007). Are tip-of-the-tongue states universal? Evidence from the speaker of an unwritten language. Memory, 15, (2), 167-176.
Brown, A. (1991). A review of the tip-of-the-tongue experience. Psychological
Bulletin, 109, 204-223.
Brown, R., & McNeill, D. (1966). The ”tip-of-the-tongue” phenomenon. Journal of
Verbal Learning and Verbal behaviour, 5, 325-579.
Burke, D. M., MacKay D. G., & James, L. E. (2000). Theoretical Approaches to Language and Aging. In T. Perfect & E. Maylor (Eds.), Models of cognitive
aging (pp. 204-237). Oxford, U.K.: Oxford University Press.
Burke, D. M., MacKay, D. G., Worthley, J. S., & Wade, E. (1991). On the tip of the tongue: What causes word finding failures in young and older adults? Journal
of Memory and Language, 30, 542-579.
Butterworth, B. (1982). Speech errors: Old data in search of new theories. In A. Cutler (Ed.), Slips of the tongue. Amsterdam: Mouton.
Cohen, G., & Faulkner, D. (1986). Memory for proper names: Age differences in retrieval. British Journal of Developmental Psychology, 4, 187-197.
Edeltrud, M. (1999). Gender processing in speech production: Evidence from German speech errors. Journals of Psycholinguistic Research 28 (5): 601-622.
Garnham, A., Shillcock, R.C., Brown, G.D.A., Mill, A.I.D., & Cutler, A. (1981). Slips of the tongue in the London-Lund corpus of spontaneous conversation. In A. Cutler (Ed.), Slips of the tongue and language production (pp. 251-263). Amsterdam: Mouton.
Harley, T. A. (1984). A critique of top-down independent levels models of speech production: Evidence from non-plan-internal speech errors. Cognitive Science,
8, 191-219.
Harley, T. A., & MacAndrew, S. B. G. (1995). Interactive models of lexicalization: Some constraints from speech error, picture naming, and neuropsychological data. In Levy, D. Bairaktaris, J. Bullinaria & D- Cairns (Eds), Connectionist
Models of Memory and Language, pp. 311-331. London: UCL Press.
Harley, T. A., & MacAndrew, S. B. G. (2001). Constraints Upon Word Substitution Speech Errors. Journal of Psycholinguistic Research, 30, (4), 395-418.
Harley, T. A., & Bown, H. E. (1998). What causes a tip-of-the-tongue state? Evidence for lexical neighbourhood effects in speech production. British Journal of
Heine, M., Ober, B., & Shenaut, G. (1999). Naturally occurring and experimentally induced tip-of-the-tongue experiences in three adult age groups. Psychology
and Aging, 14, 445-457.
James, W. (1890). The principles of psychology, 1. New York: Holt.
James, L. E., & Abrams, L. (2002). Invited book review of Schwartz, B. L. (2002) Tip-of-the-tongue states: Phenomenology, mechanism, and lexical retrieval.
The Quarterly Journal of Experimental Psychology, 58A, 573-576.
James, L. E., & Burke, D. M. (2000). Phonological Priming Effects on Word Retrieval and Tip-of-the-Tongue Experiences in Young and Older Adults.
Journal of Experimental Psychology, 26 (6), 1378-1391.
Jones, G. V. (1989). Back to Woodworth: Role of interlopers in the tip-of-the-tongue phenomenon. Memory & Cognition, 17, (1), 69-76.
Jones, G. V., & Langford, S. (1987). Phonological blocking in the tip of the tongue state. Cognition, 26, 115-122.
Kohn, S. E., Wingfield, A., Menn, L., Goodglass, H., Gleason, J.B., & Hyde, M. (1987). Lexical retrieval: The-tip-of-the-tongue phenomenon. Applied
Psycholinguistics, 8, 245-266.
Koriat, A, & Lieblich, I. (1974). What does a person in a “TOT” state know that a person in “don’t know” state doesn’t know. Memory and Cognition, 2, 647-655.
Kornell, N., & Metcalfe, J. (2006). “Blockers” do not block recall during tip-of-the-tongue states. Metacognition Learning, 1, 248-261.
Levelt, W. J. M. (1989). Speaking: From Intention to Articulation. Cambridge, MA: MIT Press/Bradford Books.
Levelt, W. J. M., Roelofs, A., & Meyer, A. (1999). A theory of lexical access in speech production. Behavioral and Brain Sciences, 22, 1-75.
Meyer, A. S., & Bock, J. K. (1992). The tip-of-the-tongue phenomenon: Blocking or partial activation? Memory and cognition, 20, (6), 715-726.
Perfect, T.J. & Hanley, J.R. (1992). The tip-of-the-tongue phenomenon: Do experimenter-presented interlopers have any effect? Cognition, 45, 55-75. Reason, J. T., & Lucas, D. (1984). Using cognitive diaries to investigate naturally
occurring memory blocks. In J. E. Harris & P. E. Morris (Eds.), Everyday
Memory Actions and Absent-Mindedness (pp. 53-70). London: Academic
Press.
Roediger, H.L., Neely, J.H., & Blaxton, T.A. (1983). Inhibition from related primes in semantic memory retrieval: A reappraisal of Brown’s (1979) paradigm.
Journal of Experimental Psychology: Learning, Memory, and Cognition, 9,
478-485.
Rubin, D. C. (1975). Within-word structure in tip-of-the-tongue phenomenon. Journal
of Verbal Learning and Verbal Behaviour, 14, 392-397.
Schwartz, B. L. (1999). Sparkling at the end of the tongue: The etiology of the tip-of-the-tongue phenomenology. Psychonomic Bulletin & Review, 6 (3), 379-393. Schwartz, B. L. (2002). Tip-of-the-tongue states: Phenomenology, mechanism, and
lexical retrieval. Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc.,
Publishers.
Nationalencyklopedin ordbok (2004). Höganäs: Bokförlaget Bra Böcker AB.
Woodworth, R. S. (1938). Experimental Psychology. New York: Holt.
Yaniv, I., & Meyer, D. E. (1987). Activation and metacognition of inaccessible stored information: Potential bases for incubation effects in problem solving. Journal
Appendix
Target Words and Definitions with Frequencies (per Million)
The frequency of the target words was taken from Allén (1971). Some of the words did not occur in that corpus.
No. Target Word F Definition
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. Akademiker Staffli Deklaration Perforering Appellera Facit Abdikera Feromoner Obduktion Reprimand Celibat Postulera Likvidation Kapsejsa (11) (2) (31) (2) (5) (3) (1) (n/a) (1) (1) (1) (2) (n/a) (n/a)
Personer som studerar eller har utbildats vid högskola eller universitet.
Trebent träställning på vilken målare ställer upp tavlor under arbetet.
Skriftlig sammanfattning av årlig inkomst- och förmögenhetsuppgifter, som lämnas in till myndighet som grund för beskattning enligt bestämda regler och på särskilda blanketter.
Rad av små hål i t ex papper för att underlätta separering av en del av papperet.
Vädja från lägre till högre domstol.
Samling lösningar till motsvarande problem. Avsäga sig tronen.
Doftämnen som överförs mellan individer av samma djurart och har ett visst budskap.
Öppning och undersökning av död kropp, vanligen för fastställande av dödsorsaken.
Formell tillrättavisning.
Avhållsamhet från sexuellt umgänge.
Förutsätta såsom sann eller existerande utan bindande bevis eller dylikt.
Frivillig eller tvångsmässig avveckling av bolag eller affärsrörelse utan konkurs.
15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Inkompetens Pacifism Amalgam Extrovert Rekognosera Implikation Chassi Arbitrage Kannibal Regression Blasfemi Friktion Kondoleans Ambivalens Hallucination Vidimera (1) (4) (n/a) (n/a) (n/a) (2) (1) (n/a) (1) (n/a) (3) (1) (n/a) (2) (2) (2)
Bristande kvalifikationer för en viss uppgift. Rörelse eller ideologisk ståndpunkt som tar avstånd från alla former av militärt våld, även då det används i försvarssyfte.
En relativt mjuk kvicksilverlegering som främst förr användes som tandfyllning.
Person som har en öppen, utåtriktad läggning. Ge sig ut i terräng för att spana och inhämta kunskap. Logisk nödvändig följd.
Underrede till bil, på vilket kaross eller lastflak monteras.
Inköp och försäljning av främmande valutor i syfte att utnyttja kursskillnader.
Person som äter människokött för att överleva, eller av rituella skäl.
Återgång till tidigare utvecklingsstadium och därmed ett barnsligare beteende.
Avsiktlig skymf eller hädelse av något som anses heligt eller värdefullt.
Det motstånd i kontaktytan som måste övervinnas när två kroppar glider mot varandra.
Formellt, högtidligt beklagande av någons sorg vid dödsfall.
Samtidigt uppträdande av två motstridiga impulser eller känslor som leder till obeslutsamhet.
Sinnesintryck, främst i form av synvillor och
hörselintryck, utan yttre motsvarighet i verkligheten, då dessa helt och hållet skapas av hjärnan.
