asp.net core pdf editor : Convert pdf back to powerpoint Library application class asp.net azure .net ajax hensonetal_qjepa_030-part1227

D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Selective interference with verbal short-term
memory for serial order information:
Anew paradigm and tests of a timing-signal
hypothesis
Richard Henson, Tom Hartley, and Neil Burgess
Institute of Cognitive Neuroscience, University College London,
London, UK
Graham Hitch and Brenda Flude
Lancaster University, Lancaster, UK
Manyrecentcomputationalmodelsofverbalshort-termmemorypostulateaseparationbetween
processessupportingmemoryfortheidentityofitemsandprocessessupportingmemoryfortheir
serialorder.Furthermore,someofthesemodelsassumethatmemoryforserialorderissupported
byatimingsignal.Wereportanattempttofindevidenceforsuchatimingsignalbycomparingan
“item probe” task,requiring memory for items, witha “list probe” task,requiring memory for
serialorder.Fourexperimentsinvestigatedeffectsofirrelevantspeech,articulatorysuppression,
temporal grouping, and paced finger tapping on these two tasks. In Experiments 1 and 2,
irrelevantspeechandarticulatory suppressionhada greaterdetrimentaleffect on thelist probe
taskthanonthe itemprobetask.Reactiontimedataindicatedthatthelistprobe task,butnotthe
itemprobetask,inducedserialrehearsalof items.Phonologicalsimilarityeffectsconfirmedthat
both probe tasks induced phonological recoding of visual inputs. Experiment 3 showed that
temporalgroupingofitemsduringlistpresentationimprovedperformance onthelistprobetask
morethanontheitemprobetask.InExperiment4,pacedtappinghadagreaterdetrimentaleffect
onthe list probe task thanon the itemprobe task. However,there wasnodifferentialeffect of
whethertappingwastoasimpleoracomplexrhythm.Overall,thedataillustratetheutilityofthe
itemprobe/list probe paradigm andprovide support for modelsthat assume memoryforserial
orderandmemoryforitemsinvolveseparate processes.Resultsaregenerallyconsistentwiththe
timing-signalhypothesis but suggest further factors that need to be explored to distinguish it
fromother accounts.
RequestsforreprintsshouldbesenttoRichardHenson,InstituteofCognitiveNeuroscience,UniversityCollege
London,17QueenSquare,LondonWC1N3AR, UK.Email:r.henson@ucl.ac.uk
ThisworkwassupportedbyaBBSRCprojectgranttoNBandGH.RHissupportedbytheWellcomeTrust.NBis
supportedbytheMedicalResearchCouncil.WethankNelsonCowan,MurrayMayberyandananonymousreferee
for theirhelpful comments.
GrahamHitchisnowattheDepartmentofPsychology,University ofYork,York YO105DD,UK.
2003 TheExperimentalPsychologySociety
http://www.tandf.co.uk/journals/pp/02724987.html
DOI:10.1080/02724980244000747
THEQUARTERLYJOURNALOFEXPERIMENTALPSYCHOLOGY, 2003,56A(8),1307–1334
Q0924—67/01
Convert pdf back to powerpoint - Library application class:C# Create PDF from PowerPoint Library to convert pptx, ppt to PDF in C#.net, ASP.NET MVC, WinForms, WPF
Online C# Tutorial for Creating PDF from Microsoft PowerPoint Presentation
www.rasteredge.com
Convert pdf back to powerpoint - Library application class:VB.NET Create PDF from PowerPoint Library to convert pptx, ppt to PDF in vb.net, ASP.NET MVC, WinForms, WPF
VB.NET Tutorial for Export PDF file from Microsoft Office PowerPoint
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Several computational models of verbal short-term memory have been developed recently
(e.g., Brown, Preece, & Hulme, 2000; Burgess & Hitch, 1992, 1999; Henson, 1998; Page &
Norris, 1998). These models go beyond previous theories, such as the phonological loop
(Baddeley,1986;Baddeley&Hitch,1974),bysimulatingserialpositioncurves,errorpatterns,
and effects of characteristic variables such as phonological similarity, word length, and
presentation modality. Moreover, the models promote further theorizing and empirical
predictions. For example, the models of Burgess and Hitch (1999) and Brown et al. (2000)
postulatethe existence ofatimingsignalthatis used to representserialorder.These models
assumethatthecodingoforder information isseparate fromthe codingofitem information,
consistentwith previous research (e.g., Bjork& Healy, 1974; Healy, 1974;McNichol, 1971;
Murdock,1976). AccordingtoBurgess and Hitch(1999)and Brown etal. (2000),the timing
signalderivesfromasetofinternaloscillators andenablesthe codingofthepositionsofitems
withinasequence(forwhich thereisconsiderableevidence,Henson,1999a).Moregenerally,
the involvement of oscillators in the encoding and retrieval of verbal material can help to
explainerrorpatterns in tasks involvingphonologicaloutput(Hartley,1995,2002;Hartley&
Houghton,1996;Vousden,Brown,&Harley,2000).However,thoughtheconceptofatiming
signalhasprovedusefulinexplainingsomeaspectsofshort-termmemory,includingeffectsof
temporal grouping (Burgess & Hitch, 1996), temporal distinctiveness (Brown et al., 2000),
andrelativepositionalcoding(Henson& Burgess,1997),directempiricalevidenceis lacking.
We report a first attempt at finding such evidence, by looking for variables that affect the
operation ofthehypotheticaltimingsignal.Tosupporttheclaimthateffects werespecificto
serialordering,we devised two probed recall tasks that differed in the degree towhich they
required maintenance of serialorder.
Item probe task
Intheitemprobe(IP)task,participantsseealistofitemspresentedsequentially,followedbya
singleprobeitem,andjudgewhetherornottheprobeitemwasinthelist(Figure1a).Thistask
waspioneered bySternberg(1969),whosuggestedthatperformancewas basedonexhaustive
serialscanningofthelistitemsinsearchoftheprobeitem.Thisconclusionwasbasedchiefly
onthefindingthatreactiontimes (RTs)increased linearlywith listlength.Subsequentdata,
however,havedisputedthisclaim(e.g.,Ashby,Tein,&Balakrishnan,1993;Baddeley&Ecob,
1973; Monsell, 1978). More sophisticated analyses by McElree and Dosher (1989) showed
that performance is better explained by direct access than by serial scanning. In fact, their
favoured modeloftheIP taskwas onebased on decaying“strengths”ofitemrepresentations
(e.g.,Wickelgren &Norman,1996),consistentwith therecencyeffectfound in accuracyand
RT as a function ofprobe position. Providingpresentation rates are reasonablyfastand the
retention interval is minimal (to minimize explicit rehearsal, Monsell, 1978), the IP task
therefore seems likely to index short-term memory for item information in the absence of
serialscanning or rehearsal.
List probe task
Weconsideredvariousproberecognitiontasksthatmightrequire accesstoserialorderinfor-
mation.Themostobviousis arelativeorder task(e.g.,aprobe“FK”promptingwhether “F”
1308
HENSON ETAL.
Library application class:How to C#: Set Image Thumbnail in C#.NET
PDF, C#.NET convert PDF to svg, C#.NET convert PDF to text, C#.NET convert PDF to images VB.NET How-to, VB.NET PDF, VB.NET Word, VB.NET Excel, VB.NET Back Color.
www.rasteredge.com
Library application class:VB.NET Word: Word Conversion SDK for Changing Word Document into
completed. To convert PDF back to Word document in VB.NET, please refer to this page: VB.NET Imaging - Convert PDF to Word Using VB.
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
and“K”wereinthesameoradifferentorderinthesequence).However,thereisevidencethat
this taskcan alsobeperformed on thebasis oftherelativelevels ofitem strengthin memory,
and possibly backward search, and so does not require maintenance of serial order per se
(Hacker,1980;McElree& Dosher,1993).Analternativemethodisaposition-itemprobetask
(e.g., a probe “F3” prompting whether or not “F” occurred in the third position of the
sequence).However, this task suffers from the additionalprocessingrequirementtodecode
thenumericrepresentationofposition.A morepromisingalternativeis alistprobe(LP)task,
similartotasksused previouslybyAllport(1984;seealsoGathercole,Service,Hitch,Adams,
&Martin, 1999).
In the LP task, a listis presented sequentially as in the IP task, but the probeis a second
simultaneously presented list that participants judge as the same or different from the first
(Figure 1b). The probe list always contains the same items as the original list and, when it
differs,itdoes soonlyin the transposition oftwoadjacentitems. We assumed thatthe probe
format would encourage forward serial processing, such that participants would compare
successive items in the probeagainsttheir memoryfor the list.Thus, whereas the IP task is
primarily a test of item information, the LP task is primarily a test of order information
(Murdock,1976).Moreover,accordingtothe“oscillatormodels”ofBurgessandHitch(1999)
andBrownetal.(2000),theLPtaskshouldinvolveutilizationofatimingsignalwhereastheIP
task should not.
Wehavealreadyusedthesetasksinafunctionalneuroimagingstudyinanattempttoisolate
brain regionsassociated with storage,rehearsal, and temporalgroupingin verbal short-term
memory (Henson, Burgess, & Frith, 2000). Relative to the IP task, the LP task activated
regions that included the left dorsal premotor cortex. Furthermore, a temporally grouped
versionoftheLPtaskproducedlessactivationinpremotorcortexthananungroupedversion.
Dorsalpremotorcortexhas been activatedpreviouslywhencomparingsequentialwithrepet-
itive finger movements(Catalan, Honda,Weeks,Cohen, & Hallet, 1998)and damagetothis
region impairs reproduction ofrhythmic motor sequences (Halsband,Ito,Tanji, & Freund,
SHORT-TERM MEMORY INTERFERENCE
1309
Figure 1. Schematicof(a)itemprobe(IP)and(b)listprobe(LP)tasks.
Library application class:C# PDF Page Rotate Library: rotate PDF page permanently in C#.net
control, RasterEdge XDoc.PDF, is a 100% clean .NET solution for C# developers to permanently rotate PDF document page and save rotated PDF document back or as
www.rasteredge.com
Library application class:C# Image: Tutorial for Collaborating, Marking & Annotating
Besides, more annotations can be drawn and saved back to the database We are dedicated to provide powerful & profession imaging controls, PDF document, image to
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
1993). Left dorsal premotor cortex may also be recruited in the perception of rhythmic
auditory stimuli whose temporal properties (amplitude modulation) are consistent with
speech (Giraudetal.,2000).Theseimagingdataare thusconsistentwith arolefor leftdorsal
premotor cortex in the processingofserialorder and rhythm,makingthis region a plausible
site for the timing signalassumed byoscillator models.
Ourassumptionbehindthepresentseriesofexperimentswas thatanyvariablethataffects
theoperation of thetimingsignal willhavea larger effect on the LP than on the IP task.We
chosetoinvestigateadiversesetofvariablesthateach possessed atemporalcomponentwhile
differing in other ways. These were irrelevant sound, articulatory suppression, temporal
grouping, and rhythmic finger tapping. If a common timing signal is responsible for
maintainingmemoryforserialorder,wewouldexpecttofindasimilarpatternofeffectsacross
all four variables. Experiment1 examined theeffect ofirrelevant sound.
EXPERIMENT 1
Irrelevant sound
The presence ofbackground sound, which people aretold toignore, can nonetheless impair
short-term memory. This effect holds whether the sound is irrelevant speech (Salamé &
Baddeley, 1982), or even simple tones, provided the sound exhibits some degree of change
overtime(Jones&Macken,1995).Oneargumentforthis“irrelevantsound effect”reflecting
more than simple attentional distraction, and the reason for its present interest, is that the
effect is generally found only for tasks that require memory for serial order. For example,
Salamé and Baddeley(1990) found an irrelevantsound effectwith serial recall,but notfree
recall,andJonesandMacken(1993)foundtheeffectwithserialrecall,butnotwith amissing-
item task. For this reason, proposals have been made that irrelevant sound interferes with
orderinformation,seenvariouslyastheformationofinter-objectlinks(Jones,1993),codingof
thepositionofitemswithasequence(Henson,1998),ortherelativeorderofitemstrengthsin
memory(Page & Norris, 1998).Alternatively,in thepresentcontext,irrelevantsound might
disrupt atimingsignal.
Nonetheless, LeCompte (1996) recently reported effects of irrelevant sound in recog-
nition,paired-associatelearning, and free recall: tasks thatprimafacie donotrequire main-
tenanceofserialorder.Thisobservationisconsistentwiththeviewthatirrelevantsoundalters
representations of item content but does not affect information about serial order, as in
Neath’s featuremodel(Neath,2000). However, though theinstructionsforthetasks used by
LeCompte (1996) do not require attention to serial order, participants may nonetheless
rehearse items serially, in an attempt to aid retention. When Beaman and Jones (1997)
minimized the use of such rehearsal strategies by requiring participants to suppress
articulation, irrelevant sound had little effect on recognition or paired-associate learning.
However, one problem with the tasks used by LeCompte,Beaman,Jones, and colleagues is
that they differ along several dimensions besides the degree of seriality involved, thus
precludingdirectcomparisons.BecausethepresentIPandLPtasksinvolvethesamemethod
oflistpresentation and yes–noresponding, performance levels arecommensurable, and the
specificity of the irrelevant sound effect to serial order can be tested directly via a task by
irrelevant-sound interaction.
1310
HENSON ETAL.
Library application class:C# TIFF: Merge and Split TIFF File(s) with C# Programming
C#.NET Demo Code for Merging & Splitting TIFF File(s). // split TIFF document into 2 parts and save them back to disk TIFFDocument.SplitDocument(sourceFilePath
www.rasteredge.com
Library application class:RasterEdge Product Refund Policy
the first step for you is to sign and send back RasterEdge Software We are dedicated to provide powerful & profession imaging controls, PDF document, image to
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Method
Participants
The20volunteersreplyingtoadvertsintheUCLPsychologyDepartmentconsistedof12menand8
women(meanageof 27.6years),andtheywerepaidforparticipating.
Materials
Listsof 5,6,or 7letterswere generatedbyrandomselectionwithout replacement from the conso-
nants BDGPTVHMQRYZ. Half of the items were phonologically confusable (BDGPTV), and half
were phonologically nonconfusable (HMQRYZ). Four blocks of 60 lists were constructed for each
participant,with16 listsoffive items,20listsof sixitems,and24 listsof sevenitems.
Half of the lists were followed bya positive probe and half by a negative probe. Forthe IP task,a
positiveprobe wasanitemfromthe list,whereasanegative probewasavocabularyitemthatwasnot in
thelist.Eachposition1toN – 1inlistsofNitemswasprobedpositivelytwice.FortheLPtask,apositive
probe wasaprobe list thatmatchedthestimuluslist inorder,whereasa negative probewasa probe list
thatdidnot.Eachposition1toN– 1wasprobednegativelytwice,the probeforpositionibeingaprobe
listinwhichitemianditemi+1were transposed.
Stimuli and responses were controlled by an IBM PC,using software written by R. Henson and
S. Zielinski at the UCL Psychology Department. The irrelevant speech consisted of 10 sentences,
spoken by different American voices, obtained from the Haskins Laboratory webpage (with kind
permissionofR.Remez).Thesentenceswere repeatedinfixedorderandplayedthroughheadphones.
The volume was set reasonably loud, though participants could reduce the volume if they found it
uncomfortable.Theprecisevolumewasnotthoughtrelevant,giventhatsoundlevelshavenodetectable
effect onthemagnitude of the irrelevantsoundeffect (Tremblay&Jones,1999).
Procedure
Eachconsonantwaspresentedinthe centreofaVDU,0.5incheshigh,at arateofoneevery750 ms
(500 mson,250msoff).Participantswereinstructedtoreadeachletterinsilence.Eachlistbeganwitha
warningsignal(“!”) andwasfollowedby arecallcue (“?”)presentedinthe same mannerastheletters.
Afterthecue,theprobeappearedcentre-screen,where itremaineduntilparticipantspressedoneoftwo
keysmarked“y”and“n”.Trialswere self-pacedbythe space-bar.Instructionsemphasizedbothspeed
andaccuracy.Inthe LPtask,participantswere alsotoldthat,iftheprobelistdifferedfromthestimulus
list, it wouldonly ever differ in the order of two adjacent items.Inthe irrelevant-soundconditions,
participantswereinstructedtoignorethe backgroundspeechandwere reassuredthattheirmemoryfor
the speechwouldnot be tested.
The order of lists within blocks was randomized, such that list length on any given trial was
unpredictable. This was to discourage participants from grouping the items (Henson, 1996; see also
Experiment3).The orderofirrelevantsoundconditionswascounterbalancedacrossparticipants,with
the constraint that the first andsecond,and third andfourthblocksinvolvedthe same probe task (to
minimize confusion between tasks). Participants received 10 practice trials, andthe experiment took
approximately 30 minutes.
Data analysis
Serial position effects were analysed inrepeated measures analyses of variance (ANOVAs) on the
arcsinofproportioncorrectandthelogarithmofRT.RTsoutsidetherange200–9000 mswereremoved
SHORT-TERM MEMORY INTERFERENCE
1311
Library application class:C# PDF: Start to Create, Load and Save PDF Document
can use PDFDocument object to do bulk operations like load, save, convert images/document to page in the document), you can save it back to a PDF file or
www.rasteredge.com
Library application class:C# Imaging - Linear ITF-14 Barcode Generator
Y to control barcode image area on PDF, TIFF, Word 14 barcode image fore and back colors in BarcodeHeight = 200; barcode.AutoResize = true; //convert barcode to
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
from analyses; missing RT values were set to the mean for that cell. Nonsphericity in ANOVAs was
accommodatedby Greenhouse–Geisser correction.
Results
The proportions of correct responses in each condition are shown in the upper panel of
Figure 2 (chance = .5). Though irrelevant sound impaired performance on both tasks, the
impairment on the LP task (M= 0.08, SD = 0.09) was greater than that on the IP task
(M= 0.02, SD = 0.03). Indeed, a 2 (irrelevant sound) × 2 (probe task) × 2 (task order)
ANOVA showed a significant interaction between irrelevant sound and probe task,
F(1, 18)= 8.97,p < .01.Noeffectsoftaskorderapproachedsignificance,Fs < 1.43,p > .25.
1312
HENSON ETAL.
Figure2. Overallperformance(upperpanel)withandwithoutirrelevantspeech(Experiment1),andperformance
whencontrolperformanceinIPandLPtasksequatedbydifferentlistlengths(lowerpanel).Errorbarsshowstandard
errorofmean(betweenparticipants).
Library application class:How to C#: Create a Winforms Control
pages, VB.NET comment annotate PDF, VB.NET delete PDF pages, VB.NET convert PDF to SVG. VB.NET How-to, VB.NET PDF, VB.NET Word, VB.NET Excel, VB.NET Back Color.
www.rasteredge.com
Library application class:VB.NET Image: VB.NET Codes to Add Antique Effect to Image with .
a touch of history to the image which can help bring back the sweet We are dedicated to provide powerful & profession imaging controls, PDF document, image to
www.rasteredge.com
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Posthoct-testsshowedsignificanteffectsofirrelevantsoundonboththeIPtask,t(20) = 2.77,
p< .05,and theLP task,t(20) = 3.95, p < .01.
One might argue that the above interaction reflects a ceiling effect, given that control
performancein the IP taskwas over 90%correctand in the LP task(85%)wassignificantly
less, t(20) = 3.81, p < .01. This argument was examined by comparing six-item lists in the
LPtaskwith seven-itemlists in theIP task (Figure2, lower panel).Controlperformance on
the two tasks no longer differed significantly, t(20) = 0.02, p = .98, and yet a two-way
ANOVAstillshowedaninteractionbetweenprobetaskandirrelevantsound,F(1, 19) = 9.54,
p< .01.
Phonological similarity and probe type effects
Totestwhetherbothprobetaskswereaccessingphonologicalshort-termmemory,perfor-
mance on confusable and nonconfusable probes was compared.For theIP task, aconfusable
probe was one in which there was at least one other item in the list that was phonologically
similar to the probe (see Materials section). For the LP task, aconfusable probe was one in
whichthetransposedletterswerephonologicallysimilartoeachother(seeFigure3).Approx-
imatelyonehalfofIPtrialsandonequarterofLPtrialscontainedaconfusableprobe.Analyses
were performedon thearcsintransformofthe proportions ofconfusable and nonconfusable
trials that had correct responses. It is important to emphasize that interest was focused on
phonologicalsimilarityeffectswithineachtask.Giventhedifferentdefinitionsofsimilarity,it
is notmeaningful tocompareeffects across tasks.
The proportions of responses for confusable and nonconfusable positive and negative
probesareshowninFigure4.IntheIPtask,asmalldisruptiveeffectofphonologicalsimilarity
SHORT-TERM MEMORY INTERFERENCE
1313
Figure 3. Example probe types and probe confusabilities. Pos =positive probe; Neg =negative probe;
Con=confusableprobe;Non =nonconfusableprobe.
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
was apparent. This was confirmed by a 2 (irrelevant sound) × 2 (probe type) × 2 (probe
confusability)ANOVA, which showed asignificanteffectofprobe confusability, F(1, 19) =
7.79,p < .05.Thephonologicalsimilarityeffectappeared larger fornegativethanforpositive
probes,though the interaction only approached significance, F(1,19) = 3.21, p = .09.
For the LP task, phonological similarity exerted a large effect in both control and irrel-
evant-sound conditions,again mainlybyreducing performanceon negative probes.Indeed,
an ANOVA showed a significant interaction between probe confusability and probe type,
F(1, 19)= 18.76, p < .001. In addition, the interaction between probe type and irrelevant
sound approached significance, F(1, 19)= 3.88, p = .06, suggesting that irrelevant sound
produced agreater disruption for positivethan for negativeprobes.
Serial position effects
Initialexamination ofserial position curves for IPand LP tasks did notsuggestany inter-
action between probeposition and irrelevantsound.Datawere thereforecollapsedacross the
irrelevant-sound manipulation in order to increase the numbers of observations.
Percentage correct as a function of probe position for each probe task and list length is
shown in the upper panel of Figure 5. For the IP task, Positions 1–6representpositions of
positive probes, and Position 0 refers to negative probes. For the LP task, Positions 1–6
represent positions of negative probes, and Position 0 refers to positive probes. Data from
Position 0 were excluded from all analyses of position effects. The IP task showed recency
and primacy effects, whereas performance in the LP task generally decreased with probe
position. This was confirmed by significant interactions between probe task and probe
position in seven-item lists, F(4.1, 77.7) = 2.43, MSE= 0.16, p = .05, and six-item lists,
F(3.1, 58.8) = 3.87, MSE= 0.15, p < .05, though not five-item lists, F(2.6, 49.5)= 1.54,
MSE= 0.13,p = .22(thelatter probably reflecting ceiling effects).
1314
HENSON ETAL.
Figure 4. Accuracy as a function of probe type and probe confusability with and without irrelevant speech
(Experiment1).IS=Irrelevantsound. SeeFigure3formoredetails.
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Themean of the median correctRTs across participants is shown as a function of probe
position in thelower panel ofFigure 5. Responses in theLP task were generallyslowerthan
thoseintheIPtask,butwhereastheIPtaskshowedlittleeffectofprobeposition,theLPtask
showedageneralincreaseinRTacrossPositions1–6.Thiswasagainconfirmedbysignificant
interactions between probe task and probe position, this time for all list lengths, Fs > 5.15,
MSE < 0.01, p < .001.The coefficients ofthebest fittinglinear regressionsofRTon probe
positionforlistlengths5–7were259,230,and141 ms/position,respectively,fortheLPtask.
RTs to positive probes (Position 0) in the LP task were approximately equal tothe average
RTs for negative probes.
SHORT-TERM MEMORY INTERFERENCE
1315
Figure 5. Accuracy and reaction times (RT)as a function ofprobepositionfor each probe task andlistlength
collapsed across irrelevant speech(Experiment 1). Position0 reflects negative probes for the IPtask and positive
probes fortheLPtask.
D
o
w
n
l
o
a
d
e
d
B
y
:
[
I
n
g
e
n
t
a
C
o
n
t
e
n
t
D
i
s
t
r
i
b
u
t
i
o
n
]
A
t
:
1
0
:
2
8
5
M
a
r
c
h
2
0
0
8
Discussion
There was a greater detrimental effectofbackground speech on the LP taskthan on the IP
task,aspredictedbythehypothesisthatirrelevantsoundisparticularlydisruptiveoftasksthat
requiremaintenanceofserialorder.Theseresultsbolsterthoseofpreviousstudies(Beaman&
Jones, 1997; Jones & Macken, 1993; Salamé & Baddeley, 1990), and go beyond them by
demonstratingasignificantinteractionbetweenirrelevant-soundandmemorytasksthatwere
matched closelyinrespects other thanthe requirementforserialorder.Theobservation that
both tasks were performed less accurately when probes were confusable provides useful
confirmation thattheywereaccessingphonologicalshort-termmemory.Thefindingthatthe
additionalerrorsinducedbybackgroundspeech intheLPtaskweremainlytopositiveprobes
suggests thatirrelevantsound increasestranspositionsinthe order ofitems in memory.Such
errors would cause participants to make incorrect “no” responses to list probes that
nonetheless matched the original sequence, whereas errors thathappened to counteract the
transposition in anegative listprobe would be far less likely.
Independentsupportfor our hypothesis thattheLPand IPtasks differintheir serial order
requirementscomesfromtheanalysisofRTsasafunctionofprobeposition.TheRT–position
functionintheIPtaskwasgenerallyflat,consistentwithdirect-accesstheoriesofthistask(e.g.,
McElree& Dosher,1989).In theLP task, theRT–position function hadan average slopeof
approximately 210 ms/item, which is close to the rate of subvocal rehearsal for familiar
monosyllables(Baddeley,1986).ItisalsoclosetotheratefoundinSternberg’ssuccessornaming
task(Sternberg,1967),anothertaskbelievedtoengageserialrehearsal.Furthermore,theslopeis
tooslow tobe attributed tothevisual scanningoflistprobes fromlefttoright,asvisualsearch
ratesforlistsoflettersareoftheorderof50ms/item(Bisanz&Resnick,1978;Pashler&Badgio,
1985).Taken together, these RT analyses provide an independentmeasure of the degree of
serialrehearsalin each task, addressingthe problemraised byBeaman and Jones (1997).
It is important to note that, in addition to the interaction between irrelevant sound and
probe task, there was a small but significant residual effect of background speech on the IP
task.Oneinterpretationisthatirrelevantsoundhasmorethanoneinterferingeffect:ageneral
distraction of attention,which affects both the IP and LP tasks, and a specific disruption of
serial ordering, which affects only the LP task. An effect of irrelevant sound on any task
requiring attention and memoryis consistent with LeCompte(1996)and Neath (2000).
One puzzle concerning RTs in the LP task is that mean correct RT for positive probes
(Probe Position 0) was faster than RTs for later negative probe positions. If the LP task is
performed viaserialrehearsalfrom thestartofthesequence,participants should onlybeable
torespond “yes”toapositiveprobewhentheyhavefinished rehearsingthewholesequence.
Thatis, such responses should be at leastas slow as responses to a negative probe involving
transposition of the last two items. One possible explanation is that participants have a
tendencytoguess“yes”whenunsure(consistentwiththegreaterincidenceof“yes”responses
intheLPtask).Ifguesseswererelativelyfast,thencorrectresponsestopositiveprobeswould
be shorter than those predicted bythe serial rehearsal hypothesis.
In summary, comparison of performance on the IP and LP tasks supports the view that
irrelevantspeechisprimarilydisruptiveofmemoryfor serialorder,thoughitmayalsohavea
weaker effecton memoryforitems.TheLPandIP tasks were alsodifferentiated byRT data
consistentwith the useofsubvocal rehearsal in retrieving serialorder information.
1316
HENSON ETAL.
Documents you may be interested
Documents you may be interested