H I G H SPEED V I D E O ANALYSIS O F W A L K I N G PHONOTAXIS I N A
DIPTERAN.
M.L. Oshinskv*. A.C. Mason and R.R. Hoy.
Neurobiology and Behavior. Cornell University, Ithaca, N Y 14853-2702.
We studied the phonotactic response of freely walking flies using high speed
video analysis. Gravid female flies (Ormia ochracea, Tachinidae, Diptera) localize
singing male crickets i n the field (calling song 5 4.5 kHz) using acoustic cues. The
classical cues for sound localization such as interaural time and interaural intensity
differences are vanishingly small in the fly. The flies possess a novel mechanically
coupled hearing organ with whlch it encodes interaural mechanical time and
mtensity differences (Robert et al., J. Comp. Physiol. 1996).
I n this study, we characterized the behavior of stationary flies to near threshold
and suprathreshold synthetic cr~cketsong at 4 kHz. The intensity threshold required
for the animal to correctly orient to a sound was 55 k 6 dB SPL ("=lo). A l l animals
tested with near threshold stimuli responded by orienting toward the speaker and
then flying. A t suprathreshold intensities the animals rarely initiated flight behavior
( ~ 2 5 ) .We propose that the flies initiated flight to low intensity stimuli due to a
perceived large distance between them and the synthetic cricket song.
Flies with a single Iesioned tympanum always circle toward the intact ear in
response to sound stimuli. Flies with both tympana lesioned do not orient toward
synthetic cricket song.
Flies can correctly orient significantly better than chance to a speaker displaced
from the midline by 10' - 20". Using neurophysmlogical data from afferent
recordings, we propose that the an~malcan resolve a 100 ps interaural niechanical
time difference.
Support: NIGMS-5T32GM07469 (MLO) and NINDCD-5ROIDC00103 (RRH)
TWO TEMPORALLY AND SPECTRALLY DISTINCT AUDITORY RESPONSES BY
THE L3 NEURON MAY BE NECESSARY FOR PHONOTAXIS BY FEMALE CRICKETS
(ACHETA DOMESTICUS). G. Atkins*. H. Binaoi. B. Navia. D. Bennen. C. Bailev and J.
.%! Dept. of Biology, Andrews University, Berrien Springs, MI 49104.
L3 is an auditory interneuron in female crickets, Acheta domesticus, that exhibits
two kinds of responses to the male's calling song and may play a role in two different
aspects of phonotaxis. The immediate response encodes the syllables 01the chirp and is
temporally selective for the syllable period of the calling song of conspecific males. The
temporal selecllveness of the immediate response is age-relaled and is correlaled with
and involved in the age-related changes in phonolaxis by these females. The prolonged
response is not temporally selective and does not encode the syllables ol the chirp. The
onset of the prolonged response requires 3-5 seconds of stimulationwith a calling song, to
reach its maximum spiking rate and 7-10 seconds to decay once Ihe calling song ceases.
This response matches the frequency selectivity and threshold (50 dB at 5 kHz) of
phonotaxis (spiking rates during the prolonged response increase to a maximum wllh
increasing intensities of 5 kHz calling songs, whereas increasing intens~tiesof 16 kHz
calling songs results In Ihe suppression ol the response).
Whether the prolonged response is necessary for phonotaxis is the subject of
current experimenls. Preliminary data suggest that unattractive calling songs, such as
one with correct lemporal fealures but with a carrier frequency of 16 kHz, can be made
attractive bv, addino a 5 kHz tone between the chirps. This combination should cause
sp r ng n me r:ma, oetheen tne mmed ale rcspmse to Ine cn rp by L3 drlu I n s trim c a
pro omea resDcnse In otner ehpe' merts attract vc ca' s r 11 cn no,ce me pr I ongeo
;espoke in ~ 3can
, be made less anraclive by adding 16 kHz tones which should decrease
the prolonged response in L3. Ongoing experimenls are attempting to evaluate the
effects of two-tone stimulation on Ihe prolonged response of L3 and on phonotaxis of the
same females.
Supported by NSF IBN 92-22127 and faculty grants from Andrews University.
77.15
77.16
A POSSIBLE BEHAVIORAL ROLE FOR THE PROLONGED RESPONSE OF L3 IN THE
PHONOTACTIC THRESHOLD OF FEMALE CRICKETS (ACHETA DOMESTICUS). 11;
Bronsert. J. Stout', and G. Atkins. Biology Dept., Andrews Univ. Berrien Springs, MI
491 04.
Female crickets (Acheta domesticus) recognize and respond phonolacticallyto the calling
song (CS) of conspecific males in an age related matter with phonotact~cthresholds
dropping behveen days 2-4 following the imaginal molt.
The ascending prothoracic interneuron L1, encodes the temporal structure of male's CS
and is tuned lo its carrier frequencies (4.6 kHz). Since L l ' s response is necessary and
sufficient for phonotaxis to occur at low CS intensities (65 dB or lower) it has been
proposed that phonotactic lhresholds are regulated by Li's sensilivily lo CS. Recently we
have shown that in some 1-day-oldfemales wilh high behavioral thresholds (above 70 dB),
Ll's lhresholds could be 30 dB or more below their phonotaclic thresholds. This poor
correlation between behavioral thresholds and L1 thresholds In some females suggesls Ihe
poss~bilityof olher neurons being involved in the regulationof behavioral thresholds.
Anolher ascending prothoracic interneuron L3, has two different responses to model CS:
an Immediate response that encodes Ihe syllables of the chirp, and a prolonged response
which does not correlate with syllables of a chirp (see adjacent abstract). The threshold of
L3's prolonged response is significantly lower than the threshold of the immediale response
at 5 kHz and is typically absent at higher frequencies. Thus the tuning of the prolonged
response in 13 matches the tuning of phonotaxis for the species. Our model proposes lhat L l
and the prolonged response of L3 together, could control phonotactic thresholds. Low
phonotaclic lhresholds would require that both L1 and Ihe prolonged response of L3 be low.
If 13's prolonged response is necessary for phonotaxis then females with increased
phonotactic thresholds (1-2 days old) would have high thresholds for the prolonged
response in L3 wh~lefemales with decreased phonotactic threshold would exhibit prolonged
responses with low thresholds. Preliminary results show lhat clear prolonged responses by
L3 do not appear in young female crickets (0-2 days old).
Funded by: NSF - IBN 92-22127 and Andrews University faculty grants.
Looming-Sensitivity i n Hummingbird Hawkmoths: Neurons and Models.
M. ~ickleirl".~,N J ~rrausfeld', T ~ e h o i v s k i ~
P , ~abes', L ~ i s k o r f ' ,'ARLDN, Tucson AZ
55721 and 'The Snlk Irisrirute, Sari Diego, CA, USA
Intraccllular recordings in Mand~rcasexrn (Sphingidae, Lepidoptera) idcntiticd a
class of wide-field neuron that responded selectively to looming or recedmg stlmuh.
Clockwise and counter-clockwse rotating spirals and expanding or contracting discs
simulated loommg and ant,-looming. Both spirals and discs provide the eye with
outwardly or inwardly movmg edges, whlie the spiral s~mulateslooming or anti-looming
~nemtaininga constant area, perimeter length, and luminance on the retina. Type I cells
are activated only by the disc and not the spiral, effectively distinguishing expansion
from contract~onby measuring perimeter length. The cell class is further drvided: Type
l a neurons responded to looming and were ~nhibitedby Image size decrcasc (antilooming) whereas type I b neurons were actlvated by anti-loomlng and ~nlubuedby
loommg.
The proposed model for type 1 neurons requires them to be mutually inhib~ted,while
being fed by two systems o f retinotopically organized directional insensitive and motlon
sensitive edge-detectors through an intermediate level of elements that elther preserve or
invert the signal. The first o f the two systems provides an excitatory output on the type
l a neuron, the latter inhibits the type I b neuron. As the edge of the looming stimulus
expands on the retina there is a recruitment o f sequentially stimulated edge detectors
increasing the depolarlzat~onof the class l a cell and increasing the inhib~tiono i the class
I b cell. The oppositc occurs with reversed stimulus: excitation of the class l a cell
diminishes, while the class I b cell is gradually released from inhibition. Reclprocal
inhibition occurs between the class l a and class I b neurons to prov~dethe observed
responses of excitation to looming and inhibition to anti-looming or the reversc Wc
rmplemented the proposed model comparing the performance of the real circust with the
model. The model proves to be able to simulate the essential features of the neuronal
c~rcuit.
Frmrled by rite NIH, HHMI, ONK, and the A v H SliJ?eng.
-
77.17
77.18
D E V E L O P M E N T O F S W I M M I N G B E H A V I O R IN L A R V A L PACIFIC
B O N I T O (Snrrln ckthensts). h$ B McFarlaneI*, C Reillv', D T Cnue2, E R
al
Uopklns Marme
Usu', L M Nevinl a n d S H Thompson' I B ~ o l o g ~ cSciences,
Station of Stanford U n i v ,P a c ~ f Grove,
~c
C A 93950, 2Lifc Sciences, Monterey
Bay
, Aquarium, Monterey, CA 93940
\ \ ' r a w ~n tlw prorcs, u f d e v e l . ~ p ~ ~.iiniu,icl
g
syst1,m 111w h ~ r h1,) 5h1dy t h r
.inatonl~c.ll, 1ii~~Iia111ca1
dnd belwvmral pruces~esL I I I ~ I ' ~
t Ih~
a~cLqIl lIl ~~l l l l l nu f
swmm111); 111 t&n\t 11\l11'5 R1m~to( O r ~ i ~1'1'r~ifur~1i~~,
r
I:a~iiily 5 r 0 m b r i d d ~ )
are hit;lily ~ i i ~ ~ r apcl,11:i(
l ~ r y f ~ h c ~h,irrlrtcrize.i
s
by p r ~ > l ~ ~ l ~
ft'rlud\
; e d llf
r a p d , powerful ~ w i i n m ~ q ; 1. ) ~ v c l o ~ ~Ii m
n ~i ;i h embryos,
~
I l k r the w o l l - r t u d ~ d
7cI11~1ish.n ~ tr.i~~spnrmt.
r
h i t ~ i n l ~the
k r 1~I1rnf1\11,b w ~ l t oIh3trIi will11112 r l a v
post f c r l ~ l i ~ n (dlrf)
l ~ m at 22.23% I n f,ir.t, all asprck o f bonito 1nro1nutur
dtvd1,prnent nrz ninrkrdly . ~ r r c l r r a t m n
l Iicn compared n ttlt t l i r 7rbrc~fi,li I 111.
larval \ l a p 1s 111111tt'd to d 11-12 dpi pc'riod, dur111); IVI~ILI~. ~ i ~ i mO~~ L~I Ul IsI ~ , i i ~ d
r r p r l i t ~ v c l ve\t7iutr \lercotypcd mrllor brhnv~ors.5 i 1 r l l SI. turning, r ~ l l t n g .
cmitliiunuc \ w ~ ~ ~ i n i dild
m r : 1 i o v ~ ~ i 1 1Deliav~ural
~:
change\ appear tu be
assuri.itcd w i t h specif~cs L c ~ l c l ~ ~ ~ i n ~ scl~nlil;(,\.
c u l . ~ r I n yart~cularbody 4 1 , ~ p rand
trunk s l ~ i f n c \ sIlurint: the Ialtrr 5t.lgrc of t l i r I.~rvalperiod (6.11 dyf). 11IS
pors111l~to r e m o w the s p n a l cord intact from tlw trunk, ninkin,; 1111',1siblc t ~ )
~ n v c , t ~ p ttile
r phywllul:lcnl p r o p i ~ t l c 5of r i t h r r w i g l (11
~ III~IIVOI~CLI
5p!!i.,1
iorcl neurons Slwrtly after tllc Iar\,al sta,:e, animals i i n l l t ~ r g o.I 5trthlngly r n p d
~ ~ ~ , ~ t , i n i otor th~,
~ ~ju\~e1111?
l ~ o ~ i stn~;c-ii~
~
which the
of an i n d ~ \ , t d u a l
~t
Iwliov, ~ l c r ~ w?-~to~ 7+ 1111.1 111a 12-2411 pcriod. Ju\,I'n~lc animal5 c u h ~ b mxny
Iclr;ll
,ln,l murl~liolul;~caIch.~r.lcter~rt~cs
o h s c r w d I I I t t ~ l u lIwnito
t
\Vr Iwlww
01 tile 111111110. a t~'leost~ p t t n i 1 7 ~for
1 1 p r ( ~ l o n p cd f i ~ i w l t
t l i ~ utir
t \IUL~~CS
.,,~~mn,~ng.!,ill y~clcfiruitful m51ght\ into t l i d~ c v r l o p n i ~ ~annl J neural control
u f locomotor , i c t ~ v ~ it ny f i s l l t ~ .
[Suppurtcd I,y grants froni ,\I I,\ and SSI-1
WHY A R E T O A D F I S H SONIC M U S C L E S SO F A S T ? M L Fme*, K
.
SOCIETY
FOR NEUROSCIENCE,
VOLUME24, 1998
Mallov. S Mitchell Dept Blotogy, V a Commonwealth Untversity,
&clunond, V A 23284-2012
Paned toadfish s o n ~ cmuscles, s~tuatedo n the lateral surface o f the
swtmbladder, contract at 200 Hz t o produce the boatwhstle advertisement
call. Although there has been extensive w o r k o n its anatomy and
physiology, the muscle's functional contribution t o sound generation has not
been considered sincc Skoglund (1961) determined that muscle contraction
rate generates call fundamental frequency. In this study w e electrically
stimulated the sonic nerve and inonitorcd contraction with emg electrodes, a
laser vibrometer to measure bladder displacement, and a microphone t o
relate sound t o movement. The muscle fibers are curved l i k e a diaphragm,
and a t w i t c h generates a complicated bladder displacement: the sides are
pushed in, increasing internal pressure, w h i c h then pushes the bottom of thc
bladder out. The acoustic waveform o f contraction mirrors this movement
w i t h a small rapid peak o f negative pressure (sides in) followed b y a larger
positive peak (bottom out). T h e peak o f acoustic pressure occurs at the
midpoint o f contraction when velocity is maximal and acceleration i s zero.
When stimulating w i t h trains, peak sound amplitude increases to almost 200
Hz,indicating a specialization f o r the boatwhistle, i s still obust at 400 Hz,
and does not completely tetanize at over 500 Hz. The bladder produces more
acoustic pressure per displacement at higher frequencies, indicating that
peak amplitude at boatwhistle frequencies is caused b y optimal muscle
contraction rather than bladder acoustics. Since sound amplitude i s
proportional t o bladder velocity, w e suggest that fast sonic muscles in
teleosts evolved t o produce louder sounds. Supported b y NM D C O 1083.