Content area
Utilising a cognitively demanding strategy-based priming paradigm, we recently observed that acute transdermal nicotine selectively influenced controlled semantic processing but not related-word links within semantic memory per se as reported by Holmes et al. (Int J Neuropsychopharmacol 11:389-399, 2008).
The current study employed a less cognitively demanding priming paradigm to investigate whether nicotine influences the activation/access of links within semantic memory, and if the selective nicotinic influence on controlled but not automatic semantic processing could also be observed with these more general priming procedures.
Transdermal nicotine patches (7 mg/24 h) were administered to healthy young adults in a double-blind, placebo-controlled, crossover design. The automatic priming task (n=18) had a low relatedness proportion (RP) and was presented at a short stimulus onset asynchrony (SOA), while the controlled priming task (n=18) had a high RP and long SOA.
The patterns of priming effects indicated that automatic and controlled processing were operating for the respective tasks. However, a nicotinic influence on semantic processing was not evident for either task, nor was interplay of nicotine and relatedness observed.
Together, the findings from the previous and current study suggest that an influence of nicotine on semantic processing may only emerge when effortful controlled processing is invoked. Furthermore, the findings suggest that nicotinic modulation of links within semantic memory may only be mediated by mnemonic processes.[PUBLICATION ABSTRACT]
Psychopharmacology (2010) 211:259265 DOI 10.1007/s00213-010-1886-8
ORIGINAL INVESTIGATION
Nicotine does not enhance basic semantic priming
Anna D. Holmes & Helen J. Chenery & David A. Copland
Received: 6 January 2010 /Accepted: 10 May 2010 /Published online: 16 June 2010 # Springer-Verlag 2010
AbstractRationale Utilising a cognitively demanding strategy-based priming paradigm, we recently observed that acute transdermal nicotine selectively influenced controlled semantic processing but not related-word links within semantic memory per se as reported by Holmes et al. (Int J Neuropsychopharmacol 11:389399, 2008).
Objective The current study employed a less cognitively demanding priming paradigm to investigate whether nicotine influences the activation/access of links within semantic memory, and if the selective nicotinic influence on controlled but not automatic semantic processing could also be observed with these more general priming procedures. Methods Transdermal nicotine patches (7 mg/24 h) were administered to healthy young adults in a double-blind, placebo-controlled, crossover design. The automatic priming task (n =18) had a low relatedness proportion (RP) and was presented at a short stimulus onset asynchrony (SOA), while the controlled priming task (n =18) had a high RP and long SOA.
Results The patterns of priming effects indicated that automatic and controlled processing were operating for the respective tasks. However, a nicotinic influence on semantic processing was not evident for either task, nor was interplay of nicotine and relatedness observed.
Conclusions Together, the findings from the previous and current study suggest that an influence of nicotine on semantic processing may only emerge when effortful controlled processing is invoked. Furthermore, the findings suggest that nicotinic modulation of links within semantic memory may only be mediated by mnemonic processes.
Keywords Nicotine . Semantic priming . Attention . Cognitive demand . Effort . Acute . Patches . Nonsmokers
Introduction
Nicotine can enhance various aspects of cognitive processing such as attention and memory (see Levin 2002 for review). In a recent study utilising a cognitively demanding strategy-based priming paradigm with young healthy adults, we observed that nicotine selectively influenced controlled semantic processing but not automatic semantic processing (Holmes et al. 2008). Although this nicotinic effect on controlled processing was in line with predictions, the results did not support predictions of a specific nicotine-induced processing advantage for related-word links within semantic memory. The current study investigates whether a nicotinic influence on relatedness can be evidenced with a different priming paradigm that does not necessarily demand high levels of cognitive effort. The study also examines whether a differential influence of nicotine on automatic and controlled semantic processing, as previously observed, can also be demonstrated under less complex semantic priming conditions.
The semantic priming effect refers to the finding of faster reaction times (RTs), and often greater accuracy, for target words preceded by related prime words (e.g. catdog) compared with unrelated prime words (e.g. tabledog; Meyer
A. D. Holmes (*) : D. A. CoplandThe University of Queensland Centre for Clinical Research, Brisbane, QLD 4072, Australiae-mail: [email protected]
H. J. Chenery : D. A. CoplandDivision of Speech Pathology, The University of Queensland, School of Health and Rehabilitation Sciences,Brisbane, QLD, Australia
260 Psychopharmacology (2010) 211:259265
and Schvaneveldt 1971). The terms facilitation and inhibition effects refer to positive differences (faster RTs) and negative differences (slower RTs), respectively, when priming trials are compared to neutral prime baselines (e.g. blankdog; Neely 1991). The major processing mechanisms thought to underlie semantic priming effects are the automatic activation of semantic representations, the controlled retrieval of semantic representations, and the inhibition of competitors (Gold et al. 2006). Traditionally, semantic processing mechanisms have been divided into processes that are automatic or controlled.
Automatic processing has been defined as (a) fast acting,(b) occurring without intention, (c) not open to conscious manipulation, (d) consuming little to no limited-capacity attentional resources, and (d) facilitating but not slowing the processing of targets (Neely 1991; Posner and Snyder 1975). The most common observed theory of automatic semantic processing is automatic spreading activation, in which the activation of a prime's conceptual representation leads to the automatic spread of activation to semantically/ associatively related concepts. This preactivation results in quicker processing of related targets, thereby resulting in priming (Collins and Loftus 1975).
Conversely, controlled processing does not meet the above criteria and can both facilitate target processing and slow the processing of unrelated or invalid information. Expectancy is a commonly referred to mechanism of controlled semantic processing. Expectancy involves the generation of a set of expected targets (generally related words) following prime presentation, which essentially facilitates the processing of targets within the expectancy set while slowing processing of those which are not (see Becker 1980; Neely 1991 for details on expectancy mechanisms).
In our previous study (Holmes et al. 2008), we used a strategy-based priming paradigm with participants instructed to expect target words from particular semantic categories to follow certain primes, while unexpected targets were also presented. The task is used to invoke, specifically at a long stimulus onset asynchrony (SOA), controlled semantic processing, namely expectancy. Nicotine was observed to selectively influence controlled semantic processing, as nicotinic effects were only present when the stimuli were presented at a long but not a short SOA; a long SOA is purported to provide a temporal window within which controlled processes are able to be engaged and committed (Neely 1991). Specifically, there was a dominance of inhibition effects under nicotine for conditions that did not meet the provided semantic strategies (i.e. for unexpected priming conditions) suggesting a nicotinic enhancement of controlled semantic processing or active inhibitory mechanisms. The findings were in line with the notion, derived particularly from mnemonic processing studies, of nicotine influencing
attentional and effortful processing (Rusted et al. 1998; Warburton et al. 2001).
Such mnemonic processing studies suggest that nicotine may modulate related-word links within semantic memory, at the least when processing is active or effortful. For the strategy-based priming paradigm under a long SOA, we had hypothesised a nicotine-induced processing advantage for the related-word priming condition that met the provided semantic strategy (i.e. the expected-related priming condition). The results, however, did not support this prediction, as no interplay of nicotine and relatedness was evident. In comparison to a more general priming paradigm where explicit expectancy instructions are not given (Keefe and Neely 1990), participants undertaking the strategy-based paradigm had the added load of maintaining the given strategies online, and upon prime presentation, determining if an attentional shift was required. It is possible that the complexity of the task may have subsumed any nicotinic effects on the activation/access of semantic links.
Furthermore, it remains unclear whether nicotine can influence the controlled activation/access of related-word links within semantic memory or if a nicotinic modulation of relatedness is mediated by other memory processes (e.g. encoding and/or consolidation). Extrapolating the findings of mnemonic processing studies to nicotinic effects on semantic memory generally has been confounded by methodological issues, particularly in the case of Rusted et al. (1998, experiment (Exp) 2). Nicotine reportedly enhanced the recall of strongly associated word pairs of the same grammatical class (e.g. boygirl) but not encapsulated word pairs (word pairs with a derived meaning independent from the meaning of the components, e.g. snowman) or unrelated word pairs. However, although not primary associates, the unrelated word pairs were in fact related. Moreover, the related word pairs had a mean frequency that was 4.518.2 times larger than the encapsulated and unrelated word pairs. The current study therefore sought to investigate the effects of nicotine on the activation/access of semantic memory with more general priming procedures that do not necessarily demand high levels of cognitive effort.
Aside from instruction, other factors have been shown to influence whether automatic or controlled processes are induced in a semantic priming task. For example, automatic processes are most effective when the SOA between the prime and target is short, and thus contribute to priming effects within a restricted temporal window, whereas controlled processes are employed at long SOAs (>400 ms; de Groot 1984; Neely 1977). Increasing the number of related prime-target pairs in the stimulus list, that is, increasing the relatedness proportion (RP), is another factor that has been shown to promote controlled
Psychopharmacology (2010) 211:259265 261
processing (Keefe and Neely 1990). Hence, a controlled priming task comprising a long SOA and high RP, and an automatic priming task comprising a short SOA and low RP were utilised in the present study to investigate nicotinic effects on controlled and automatic semantic processing, respectively. Prime-target category relation(i.e. category versus noncategory) was also manipulated to investigate if nicotine differentially modulates such prime-target relationships.
We previously observed an influence of nicotine on controlled semantic processing but not automatic processing (Holmes et al. 2008), a finding in line with the assertion that nicotinic effects on behavioural performance are mediated by effortful processing (Rusted et al. 1998; Warburton et al. 2001). Thus, in the current study, it was predicted that nicotinic effects would be evident for the controlled priming task but not the automatic priming task. It was further predicted, given the reduced complexity of the paradigm compared with the strategy-based paradigm together with interplay of nicotine and relatedness in mnemonic processing studies (Rusted et al. 1998, Exp 1; Warburton et al. 2001), that a nicotinic modulation of the activation/access of pre-existing links within semantic memory would be evident for the controlled priming task. Specifically, priming effects and facilitation for related conditions were predicted to be greater under nicotine than placebo. Furthermore, as nicotinic effects appear to be mediated by the processing of word pairs with a high association from the same grammatical class (Rusted et al. 1998, Exp 2), it was predicted that intracategory priming conditions (e.g. coffeetea) may demonstrate more sensitivity to nicotinic modulation than noncategory priming conditions (e.g. artistpaint) due to greater semantic relatedness or feature overlap.
Method
Design
Each of the automatic and controlled priming experiments had a 224 (patch orderdrugcondition) mixed factor design with patch order (AB, BA) as a between-subjects factor, and drug (placebo, nicotine) and condition (category-related, CR; noncategory-related, NCR; unrelated, U; neutral, N) as within-subjects factors. Under double-blind conditions, participants were randomly assigned to receive either a transdermal nicotine patch (NicabateCQ 7 mg/24 h) or placebo (identical in size and colour to the nicotine patch) in the first session with the alternate patch in the following session. Participants were also randomly assigned to the automatic (n =19) or controlled priming (n =18) tasks.
Participants
Thirty-seven healthy, right-handed, English-speaking non-smokers participated in the study. Participants for the automatic priming task (8 males, 11 females) were aged 1831 (meanSD, 24.84.3) and had an average education of 162.3 years. Participants for the controlled priming task (8 males, 10 females) were aged 2034 (26.24.4) and had an average education of 17.32.5 years. There was no significant difference for participant age (p=0.327) or years of education (p=0.118) across the tasks. Participants were excluded if they (a) had uncorrected visual impairment; (b) were unable to perform the experimental task; (c) were pregnant, lactating or planning to become pregnant; (d) had a history of drug or alcohol abuse; (e) had a history of neurological, psychiatric or language disorder; or (f) had any contraindication for the use of a nicotine patch. The study was approved by the University of Queensland Medical Research Ethics Committee, and all participants provided written informed consent.
Apparatus
Stimulus presentation and response collection were controlled and timed by E-Prime Version 1.1 (Psychology Software Tools, Inc: http://pstnet.com
Web End =http://pstnet.com ) software running on a Pentium PC with a Model 200a PST serial response box connected.
Stimuli
Each of the priming tasks included the following 48 critical priming trials: (a) 12 CR (e.g. coffeetea); (b) 12 NCR (e.g. artistpaint); (c) 12 U (e.g. flatarmy); and (d)12 N (e.g. blankhour). The CR and NCR word pairs were selected from the stimulus list of Ober et al. (1997); the two conditions were matched for association (Edinburgh Associative Thesaurus: http://www.eat.rl.ac.uk/
Web End =http://www.eat.rl.ac.uk/ ; average association was 0.26 and 0.27, respectively). For both automatic and controlled tasks, targets for all four conditions were matched for imageability, number of letters and written frequency (MRC Psycholinguistics Database: http://www.psy.uwa.edu.au/mrcdatabase/uwa_mrc.htm
Web End =http://www.psy.uwa.edu.au/mrcdatabase/ http://www.psy.uwa.edu.au/mrcdatabase/uwa_mrc.htm
Web End =uwa_mrc.htm ), and mean lexical decision RT and accuracy (English Lexicon Project: http://elexicon.wustl.edu/default.asp
Web End =http://elexicon.wustl.edu/ http://elexicon.wustl.edu/default.asp
Web End =default.asp ). Twelve neutral prime-nonword target pairs (e.g. blankyuct) and 158 word prime-nonword target pairs were created (e.g. cardbremp) for the lexical decision tasks. Nonword targets were taken from the ARC Nonword Database (http://www.maccs.mq.edu.au/<nwdb/
Web End =http://www.maccs.mq.edu.au/ http://www.maccs.mq.edu.au/<nwdb/
Web End =nwdb/ ). Nonwords were orthographically legal and pronounceable, and were not homophonic with real English words.
262 Psychopharmacology (2010) 211:259265
Automatic priming task
The automatic task consisted of six blocks of 57 trials. The automatic priming stimuli comprised the 48 critical priming trials, the 12 neutral prime-nonword pairs, the 158 word prime-nonword pairs, as well as 124 unrelated word pair fillers. Thus, the automatic priming task comprised an RP (defined as the number of related word targets divided by the number of related plus unrelated word targets) of 0.15 (24/160) and a nonword ratio (NWR, defined as the number of word prime-nonword targets divided by the total number of word prime-nonword targets plus unrelated word targets) of 0.54 (158/294). Within a block, two-word pairs from each of the critical priming conditions and two neutral prime-nonword pairs were presented. The word prime-nonword pairs and unrelated word pair fillers were distributed across the block such that four of the blocks consisted of 26 word prime-nonword pairs and 21 unrelated word pair fillers, while two blocks consisted of 27 word prime-nonword pairs and 20 unrelated word pair fillers.
Controlled priming task
The controlled task consisted of three blocks of 48 trials. The controlled priming stimuli included the 48 critical priming trials, the 12 neutral prime-nonword pairs, and 60 of the word prime-nonword pairs, as well as 24 related word pair fillers; all divided equally across the three blocks. Thus, the controlled priming task comprised a 0.8 RP (48/60) and a 0.83 NWR (60/72).
Procedure
Placebo and nicotine patches were covered with an opaque plaster and applied to the upper arm. Participants were instructed to abstain from caffeinated or alcoholic beverages, and any illicit substances 12 and 24 h, respectively, prior to patch application. Patches were applied at 09:30 hours, and experimental sessions occurred 4 h after patch application at 13:30 hours. Experimental sessions were 710 days apart. Participants completed a short practice prior to the experimental sessions with stimuli of a similar construction to that used in experimental sessions.
Each experimental trial began with a plus sign presented in the centre of the screen for 500 ms followed by 1,000 ms of blank screen. The first prime word then appeared for either 150 ms or 950 ms (automatic and controlled priming tasks, respectively), followed by 50 ms of blank screen. The target word followed immediately, remaining on the screen for a maximum of 2,000 ms. Participants were instructed to respond as quickly and accurately as possible to the second
letter-string of the pair by pressing a yes button if it was a real word or pressing a no button if it was not a real word(i.e. a nonword). The intertrial interval was 1,000 ms.
Each block commenced with five buffer trials.
Subsequent trials within a block were pseudo-randomised so that the same condition or yes/no response never occurred more than three times consecutively. Block presentation was randomised. The experimental block was preceded by 10 practice trials. The response measure of interest was the RT for correct responses to real word targets.
Results
One participant in the automatic priming task was identified as having a high error rate (>10%) and was excluded from the analysis. With this participant removed, the overall error rates for real word trials were 1% and 2% for the automatic and controlled priming tasks, respectively, and the nonword error rate was 4% for both tasks. Due to the low percentage of errors, no further analyses were conducted on the error data. RT data for critical trials were excluded if <100 ms or >1,000 ms, and if more than 2 SD above or below the condition mean per participant. RTs >2 SD above or below the condition mean per group were also removed to normalise the distribution (9.7% data removed). For each of the priming tasks, individual participant RTs for correct responses to real word targets were submitted to a linear mixed model (LMM) analysis, with patch order (AB and BA), drug (placebo and nicotine), and condition (CR, NCR, U and N) as fixed factors. Participant RT variations in addition to participant variation across testing sessions were treated as random factors.
Mean RTs and standard errors for each priming task as a function of drug and condition are shown in Table 1 (patch order, which did not have an effect, is not included for ease of review). Analyses revealed a significant main effect of condition only for both the automatic (F(3, 1,495)=8.90, p<
0.001) and controlled priming tasks (F(3, 1,493)=23.56, p<
0.001). No other main effects or interactions with drug were significant for either priming task (Fs<2, ps>0.197) for the automatic task and (Fs<1, ps>0.424) for the controlled task). Pairwise contrasts on the data collapsed across drug for the automatic priming task indicated for both the CR and NCR conditions significant priming (i.e. faster RTs compared to the unrelated condition; p=0.001 and p=0.029, respectively; Fig. 1a) and facilitation (i.e. faster RTs compared to the neutral condition; ps<0.001; Fig. 1b), with no significant inhibition for the unrelated condition. Contrasts on the data collapsed across drug for the controlled priming task indicated significant priming (ps<0.001; Fig. 2a) and facilitation (ps<0.001; Fig. 2b) for both
Psychopharmacology (2010) 211:259265 263
Table 1 Mean RT per condition as a function of drug for the automatic and controlled priming tasks
Condition
CR NCR U N
Automatic priming task
Placebo 477 (11) 482 (11) 490 (11) 500 (11) Nicotine 468 (11) 474 (11) 485 (11) 487 (11) Controlled priming task
Placebo 470 (14) 459 (14) 499 (14) 484 (14) Nicotine 469 (14) 469 (14) 499 (14) 491 (14)
Mean RT in milliseconds and SEM in parentheses
RT reaction time, CR category-related, NCR noncategory-related, U unrelated, N neutral
the CR and NCR conditions, and significant inhibition for the unrelated condition (p=0.015; Fig. 2b). There was no significant difference between the CR and NCR conditions for either the automatic (p=0.209) or controlled (p=0.232) priming tasks.
Transdermal patch tolerance
No participant withdrew from the study due to experiencing side effects from the patch. Of the 37 participants, eight
reported experiencing side effects of a mild severity: slight nausea (n=3); headache following patch application that resolved before testing (n=3); and itching at the patch site (n=4). The participant excluded from data analysis due to a high error rate did not report experiencing any side effects. Of the participants, 54.1% identified the order in which they received the patches. In order to determine if participants' subjective feelings varied with drug state, 14 of 16 items of the Bond and Lader (1974) mood rating scale were administered (excluding items Bored and Gregarious). Paired-sample t tests revealed no significant effect of drug (ps>0.05) for any item.
Discussion
We previously observed an effect of nicotine on controlled semantic processing with strategy-based priming procedures but no direct interplay of nicotine and relatedness. The current experiment investigated whether such an interaction would be revealed with more general priming procedures. As hypothesised, there was no effect of
Fig. 1 Automatic priming task: (a) priming effects and (b) facilitation/inhibition effects. CR category-related, NCR noncategory-related, U unrelated, N neutral. *p<0.05; p<0.001
Fig. 2 Controlled priming task: (a) priming effects and (b) facilitation/inhibition effects. CR category-related, NCR noncategory-related, U unrelated, N neutral. *p<0.05; p<0.001
264 Psychopharmacology (2010) 211:259265
nicotine for the automatic priming task, supporting the notion that nicotine does not influence automatic semantic processing. Contrary to predictions, however, there was no effect of nicotine for the controlled priming task. In light of previous findings of an effect of nicotine with strategy-based priming procedures, the notion that nicotine effects are not observable if effortful semantic mechanisms are not invoked will be discussed. Furthermore, the conditions under which nicotine may modulate pre-existing links and/ or their activation/access within the semantic network to enhance behavioural performance will be considered.
For the automatic priming task, significant priming and facilitation for related conditions without significant inhibition for unrelated conditions indicated the operation of automatic processing mechanisms (Neely 1991; Posner and Snyder 1975). The absence of a drug effect suggests that nicotine does not influence automatic processing and is in line with our previous findings under different priming procedures (Holmes et al. 2008) and findings from mnemonic processing studies (Edginton and Rusted 2003; Rusted et al. 1998; Warburton et al. 2001). Furthermore, the finding suggests that, in the absence of controlled processing, nicotine does not influence the spread of activation between related-word links within semantic memory.
For the controlled priming task, significant inhibition for unrelated conditions as well as significant priming and facilitation effects for related conditions suggested that controlled processing was operating (Neely 1991; Posner and Snyder 1975). However, there was no effect of drug, contrasting with our previous observations with strategy-based priming procedures (Holmes et al. 2008). The current task was employed for its comparative simplicity, and the lack of nicotine effect may be due to the noneffortful nature of the evoked processing. In comparison to the strategy-based priming procedures, the level of cognitive effort required was limited and largely implicit. Participants in the current study were not required to maintain complex heuristics online, nor were they faced with conflicting stimulus conditions. It has been demonstrated that merely maintaining information about a prime imposes a cognitive load (Smith et al. 2001). Additionally, the level of uncertainty, number of choices and amount of information conveyed by a stimulus (or prime, in this instance) directly relate to processing requirements.
Furthermore, the findings of a recent study by Hutchison (2007) suggest that encouraging the use of controlled expectancy explicitly (e.g. overt trial by trial indication of the probability that a target would be related to a prime) leads to priming effects that are due to effortful processing, while encouraging the induction of controlled processing implicitly (e.g. increasing the RP list-wide) may lead to priming effects which, in part, may be due to unconscious processing mechanisms. Such evidence provides support
for an alternative to the traditional classification of semantic processing as outlined in the introduction, to a classification of automatic and context-sensitive processing, with the latter consisting of two subclasseseffortful controlled processes and conditionally automatic processes (see Stolz et al. 2005 for review). In this framework, effortful controlled processes are strategically engaged and require relatively constant use of attentional resources. This effortful processing likely mediated the effects of nicotine under the previous strategy-based priming procedures, where participants had to maintain complex heuristics, determine if attentional shifts were required, generate expectancy sets, and possibly inhibit semantic competitors. In contrast, conditionally automatic processes operate relatively automatically or effortlessly once the right task conditions and goals or intentions are in effect. It is likely that the current controlled priming task with an implicit high RP evoked conditionally automatic processing. As such processing does not require effort and the constant service of attention, an effect of nicotine was not observed.
Warburton and Rusted (1993) suggest that, in relation to memory studies, nicotinic effects may reflect the attentional demands of the task and the effective employment of additional nicotine-derived resources. Burgeoning evidence from animal models implicates prefrontal acetylcholine efflux in mediating attentional functioning, and more specifically increases in attentional effort (Sarter et al. 2006). In such models, nicotine has been shown to enhance performance for attentionally demanding tasks (Young et al. 2004), and attentional impairments due to nicotinic deficiencies appear central to other deficits in cognition (Young et al. 2007). Thus, the lack of nicotine effect for the current task suggests that the evoked processing was relatively effortless and did not tax attentional resources.
In keeping with our previous findings, facilitated processing of related conditions under nicotine was not evidenced. Interplay of nicotine and relatedness with mnemonic paradigms but not semantic priming paradigms suggests that a nicotinic influence on relatedness per se may best be mediated by encoding and/or consolidation mechanisms. Nicotine may enhance the strength of pre-existing links following activation or assist in the maintenance of that activation (Rusted et al. 1998). Alternatively, nicotine may increase the attentional resources able to be devoted to rehearsal of these relations (Warburton and Rusted 1993). Together, the lack of a specific relatedness advantage under nicotine for the previous and current studies suggests that nicotine may not specifically assist in enhancing the initial activation of pre-existing links for related word pairs. However, findings of enhanced inhibition effects under the previous strategy-based priming paradigm (Holmes et al. 2008) and under a retrieval-induced-forgetting paradigm (Edginton and Rusted 2003) suggest that nicotine may
Psychopharmacology (2010) 211:259265 265
influence memory networks by optimising inhibitory mechanisms operating during effortful processing. In conclusion, in light of our previous findings, the null results of the current study suggest that an influence of nicotine, at least on inhibitory semantic mechanisms, may only emerge when the level of effortful controlled semantic processing is high.
Acknowledgements The authors would like to acknowledge GlaxoSmithKline for the donation of transdermal nicotine patches for the purpose of this study. David Copland is supported by a NHMRC Career Development Award. The experiments comply with the current laws of the country in which they were performed.
Conflict of interest None declared.
References
Becker CA (1980) Semantic context effects in visual word recognition: an analysis of semantic strategies. Mem Cognit 8:493512
Bond A, Lader M (1974) The use of analogue scales in rating subjective feelings. Br J Med Psychol 47:211218
Collins AM, Loftus EF (1975) Spreading activation theory of semantic processing. Psychol Rev 82:407428
de Groot AMB (1984) Primed lexical decision: combined effects of the proportion of related prime-target pairs and the stimulus-onset asynchrony of prime and target. Q J Exp Psychol A 36:253280
Edginton T, Rusted JM (2003) Separate and combined effects of scopolamine and nicotine on retrieval-induced forgetting. Psychopharmacology 170:351357
Gold BT, Balota DA, Jones SJ, Powell DK, Smith CD, Andersen AH
(2006) Dissociation of automatic and strategic lexical-semantics: functional magnetic resonance imaging evidence for differing roles of multiple frontotemporal regions. J Neurosci 26:65236532 Holmes AD, Chenery HJ, Copland DA (2008) Transdermal nicotine modulates strategy-based attentional semantic processing in non-smokers. Int J Neuropsychopharmacol 11:389399Hutchison KA (2007) Attentional control and the relatedness proportion effect in semantic priming. J Exp Psychol Learn Mem Cogn 33:645662
Keefe DE, Neely JH (1990) Semantic priming in the pronunciation task: the role of prospective prime-generated expectancies. Mem Cognition 18:289298
Levin ED (2002) Nicotinic receptor subtypes and cognitive function. J
Neurobiol 53:633640Meyer DE, Schvaneveldt RW (1971) Facilitation in recognizing pairs of words: evidence of a dependence between retrieval operations. J Exp Psychol 90:227234Neely JH (1977) Semantic priming and retrieval from the lexical memory: roles of inhibitionless spreading activation and limited-capacity attention. J Exp Psychol Gen 106:226254Neely JH (1991) Semantic priming effects in visual word recognition: a selective review of current findings and theories. In: Besner D, Humphreys GW (eds) Basic processes in reading: visual word recognition. Erlbaum, Hillsdale, pp 264336Ober BA, Vinogradov S, Shenaut GK (1997) Automatic versus controlled semantic priming in schizophrenia. Neuropsychology 11:506513Posner MI, Snyder CRR (1975) Attention and cognitive control. In:
Solso RL (ed) Information processing and cognition: the Loyola symposium. Erlbaum, Hillsdale, pp 5585Rusted JM, Graupner L, Tennant A, Warburton DM (1998) Effortful processing is a requirement for nicotine-induced improvements in memory. Psychopharmacology 138:362368Sarter M, Gehring WJ, Kozak R (2006) More attention must be paid: the neurobiology of attentional effort. Brain Res Brain Res Rev 51:145160Smith MC, Bentin S, Spalek TM (2001) Attention constraints of semantic activation during visual word recognition. J Exp Psychol Learn Mem Cogn 27:12891298Stolz JA, Besner D, Carr TH (2005) Implications of measures of reliability for theories of priming: activity in semantic memory is inherently noisy and uncoordinated. Vis Cogn 12:284336 Warburton DM, Rusted JM (1993) Cholinergic control of cognitive resources. Neuropsychobiology 28:4346Warburton DM, Skinner A, Martin CD (2001) Improved incidental memory with nicotine after semantic processing, but not after phonological processing. Psychopharmacology 153:258263 Young JW, Finlayson K, Spratt C, Marston HM, Crawford N, Kelly JS,
Sharkey J (2004) Nicotine improves sustained attention in mice: evidence for involvement of the alpha 7 nicotinic acetylcholine receptor. Neuropsychopharmacology 29:891900Young JW, Crawford N, Kelly JS, Kerr LE, Marston HM, Spratt C,
Finlayson K, Sharkey J (2007) Impaired attention is central to the cognitive deficits observed in alpha 7 deficient mice. Eur Neuropsychopharmacol 17:145155
Springer-Verlag 2010