PS1061: Sensation and Perception 2013
Term 2, Thursday 11 am - 1 pm (Boiler House)

Lecture 7: Touch, smell, taste: Basic, but hidden senses

Course co-ordinator: Johannes M. Zanker,, (Room W 246)

Lecture Topics

somatic senses

somatic senses: sensing the physical state of the body is based on a variety of information:

significance of the tactile sense

what is the purpose of touch for human behaviour ?

the infamous moment when Australian Prime Minister Paul Keating grabbed the Queen (1992), earning himself the nickname the 'Lizard of Oz’ (BBC News 2000)

signals and sensors

pressure, deformation, strain needs to be measured

for this, we have a large number of mechanoreceptors in our skin, muscles, connective tissue

different types of mechanoreceptors respond to different physical conditions

NOTE: each of these receptors is specific in their tuning for frequency, they show adaptation effects, and have receptive fields...

tactile receptive fields

two pressure points are perceived as separate if they are stimulating the receptive fields different mechanosensors (adapted from Goldstein 2002)

the receptive field size determines spatial resolution (the number of points that can be detected in a given skin area)

tactile resolution (receptive field size) varies for different areas of the body surface: finger tips are better than palm of the hand, etc...

space of touch

a psychophysical approach to test the limits of touch perception

how many points can we resolve on the skin ?

A: what is the smallest distance of two objects touching the skin that can be discriminated?

B: how well can we identify the location of an object touching the skin relative to others?

spatial resolution of touch: down to the range of millimeters on fingertips & lips, several centimeters on back (after Sekuler & Blake 2002)

sensitivity limits of touch

what is the precision of identifying / discriminating different surface textures by touching them ?

pressure amplitudes (depth variations that reflect the texture of a surface) that only span a range of a few tens of micrometers can be detected and discriminated by moving the fingertips across them !!

(after Sekuler & Blake 2002)

note that moving the fingertips across a surface is a typical case of active explration, which is a common strategy in sensory perception

cortical representation

tactile information is transmitted from the mechanoreceptors to the somatosensory cortex to be further processed (and converted into percepts) in this region

in a first step, a somatosensory map is generated : a topologically ordered representation of the body surface on the somatosensory cortex surface – this map is called the ‘homunculus

(after Penfield & Rasmussen 1950)

somatosensory representation

how is the local (fingertip, for instance) information represented in the cortex ?

the amount of space (cortical volume) in the cortical map dedicated to a given region of the skin (body surface) differs widely; the scale of the map depends on the somatosensory resolution (number of touch points per skin area– maximum for fingertips and lips)

motor representation

similar to the somatosensory representation, and closely linked to it, there is a motor map in the cortex that represents and movement patterns and is involved in motor control

(cartoon by Gary Larson)

cortical plasticity

after amputation, some patients experience a variety vivid sensations from the missing limb: phantom limbs

V. Ramachandran (2000) describes a patient who localizes individual fingers of the amputated hands on particular regions of the face -- this is interpreted as result of cortical re-organisation of sensory input from the amputation stump (and is related to the neighbourhood of the cortical representations of these regions: index finger I, thumb T, pinky P)

… active exploration …

for touch perception the active movement of the ‘sensory organ’ (skin surface, fingers, hands: haptic exploration) is crucial
it is a strategy that is continuously applied in various ways: we talk about ‘handling’ objects

(adapted from Sekuler & Blake 2002)

(NOTE: this is similar to eye, head, body movements in vision)

… and in this domain, it is possible to generate a 'haptic' Muller-Lyer illusion                  (see Heller et al. 2002)


reading by touch - Braille

Braille letters are specifically designed symbols for stimulating the fingertips, which allows blind people to read...

what is so special about Braille letters, why not just using embossed Roman characters ?

Braille is adapted to low resolution of touch:

  •   if Roman letters (e) are blurred (f), they can no longer be discriminated
  •   if Braille characters (b) are blurred (c), they still have very characteristic (discriminable) shapes !


so why not just using larger Roman characters ?
this is not favourable because of the restricted scope of fingertips: only few characters can be touched at the same time ...

‘olfactory’ perception – smell

what is the significance of smell for everyday behaviour ?

how can we classify odours ?

experimental issues: what are the lower limits of detecting odours, can we measure the intensity of smells, can we localise them ?

smell : the sensory system

the sensory organ for smell has a very intricate physical and biochemical design

the air flows through nose & mouth

it passes the olfactory epithelium in nasal cavity: full of chemosensors

sensory neurones carry information to glomeruli in olfactory bulb and from there to sensory cortex

NOTE: olfactory receptors have very narrow sensory tuning, because they are binding very specifically to chemicals !! (only a very small group of molecules would bind to a given receptor); there are about 1000 different types of receptors (Nobel Prize 2004 was awarded to Richard Axel & Linda Buck, working on these recpetors)

recognition of odours

smell: how many different odours can we identify or discriminate?

 recognition performce  for a wide range of odours

(female better than male participants: white bars; male better than female participants: black bars)


how reliable is odour discrimination ?

classification: smell space

can we identify a small basis set of fundamental odours? (simple sensory space, like brightness, primary colours, pitch)

6 primary (independent) odour qualities: fragrant, etheral, resinous, spicey, burned, putrid

they form a basis to describe any smell sensation as combination of these 6 components

(similar to trichromatic colour vision)

this corresponds to 6 types of receptors, is reflected by specific anosmias ('smell blindness')

an olfactory code ?

encoding these complex smell sensations in neural activity (reflecting perceptual classification of odours !!) is not straight-forward >>> there is a combinatorial olfactory code
(after Goldstein 2002)

little is known in humans, but population code (a specific pattern of activity that is distributed across a group of neurones reflects a given odour) is used in a wide range of biological systems …

further up in the senory processing stream, we find neuronal activity pattrens that suggest the existence of an olfactory map ...


odour localisation, sniffing

how well can we localise the source of an odour ?
(compare this to other sensory qualities, like vision, hearing, touch, ... )

experimentally, this problem is less easy to approach than in vision or hearing!

some researchers invented some ingenious sniffo-meters…

(this looks like real fun!)

NOTE: other animals seem to perform much better (think of HMC's sniffer dogs...)

(after Sekuler & Blake, 2002)

‘gustatory’ perception: taste

taste – another type of chemical sense (for chemicals in solution) …

why is it important to have a sense of taste ?

why is it an interesting scientific topic ?

taste: the sensory system

anatomiocal image of the surface of the tongue (Grey's Anatomy, 1918):

large papillae carry a huge number of chemoreceptors (taste buds)

chemosensors transmit information through gustatory neurones to the frontal cortex for further processing (including some interaction with olfactory information)

distributed coding in neurone population !!

NOTE: these sensors are exposed to many inadequate stimuli, like heat or aggressive chemicals, leading to massive cell death and continuous regeneration
<< think about the consequences for neuronal connectivity and sensory information processing! >>

fundamental taste space ?

can we identify basic taste qualities ?

4 different types of chemoreceptors (in the taste buds) sense chemical signals from bitter, sweet, sour and salty substances

traditionally they are regarded as physiological basis of 4 perceptual categories (new evidence suggests existence of a fifth taste category: ‘umami’ or 'savoury')

(after Goldstein 2002)

    •   but: ‘savoury’ discussed as fifth …
    •   experimental evidence (similar to colour vision)
        •   mixtures of categories cannot be perceived
        •   no cross-adaptation
        •   aftereffects – water tastes sweet after bitter/sour adaptation

in everyday life, we tend to use qualitative descriptions and refer to objects when describing tastes <<think of the flavour of lollies…>>

perceiving taste quality

there is little psychophysical knowledge
there is no direct spatial representation, because taste stimuli are carried by the fluid medium passing the tounge...
one example of classical knowledge about encoding of gustatory information (after Hänig 1901), (but this now questioned, see here)

the relative sensitivity for detecting    sweet - bitter - salty - sour    vary considerably for different regions of the tongue

crosstalk between senses

in general, there are very close (and often learned: Rolls et al 1996) smell-taste associations, for instance in the flavour of food !
the nature of enjoying food is relying on a combination of senses: smell - taste – texture - looks (which sensory systems are involved ? )
flavour is a complex perceptual phenomenon …

other interactions of chemical senses, like direct sensory crosstalk (in-adequate stimulatiuon on receptor level), are quite common
one example for a well investigated taste problem is hot food, crossing sensory boundaries at a low processing level

hot chili contain a chemical called 'capsaicin‘ - they activate nociceptive C fibres (pain receptors), which are also activated by heat ! (Caterina, MJ et al. 1997)

(red hot chili pepper)

NOTE: one chemical activates two sensory systems!

in reversal, capsaicin can elevate pain thresholds by adaptation of pain receptors …


summary: touch, smell, taste


Specific References:

to download a pdf copy of lecture slides, click here

back to course outline
last update 8-03-2014
Johannes M. Zanker