View Full Version : The first two of six reasons why A/B testing as conventionally applied is next to

Neil McCauley
18-10-2008, 15:52
By A/B conventionally applied I mean the same musical phrase, bar or track used to demonstrate the differences between (say) any two pairs of speakers (or DACs or whatever) where the total number of similar components on demonstration is 2 + n, where n = any number and where the process is carried out too quickly.

There are 6 key errors of judgement in the conventional method of selection of any two or more items such as cars, washing machines and so on in the audiophile field, or when choosing a one night stand or long term partner for instance in the human arena. Anyway, focusing on audiophile equipment, here are the first 2

The Halo Effect:

Observations of people’s behaviour in my demo room suggests that they perceive most pieces of equipment as possessing both appropriate and inappropriate sonic characteristics.

The individual component which is superior on all favourable characteristics is extremely rare as is the component which has no redeeming features. Observation over many years here indicates that some listeners frequently perceive the sonic performance of equipment in these black and white terms.

Under these circumstances, equipment can tend to be judged by the listener as all good or all bad. This halo effect is particularly likely to occur where a component has a single outstanding characteristic revealed in the demonstration. For example, if a component is unusually high on one attribute, (PRaT?) listeners typically tend to minimise or ignore any weaknesses they hear in other areas.

The Contrast Effect:

This is a similarly damaging and associated distortion during the demonstration. Contrast occurs when listeners fail to assure absolute standards of judgement when assessing the performance of components. Instead, the assessment of each component is influenced by the quality of the immediately preceding ones. For example, if a person hears two or three weak aspects in component ‘A’ followed by an average one in component ‘B’, the latter is often judged too leniently.

No wonder the used market is flooded by very fine equipment that turned out to be unsuitable in the sellers’ home!