TABLE 6.
The results for the case with addition of phenotypic noise
|
s of Gaussian noise
|
Position in statistics
|
|||
|---|---|---|---|---|
| Of the pattern set
|
Of the pattern (with its role sign)
|
|||
| D | E | F | ||
| 0 | 1 | 1+ | 3− | 2+ |
| 0.1 | 1 | 1+ | 3− | 2+ |
| 0.2 | 1 | 1+ | 3− | 2+ |
| 0.3 | 3 | 1+ | 2− | 3+ |
| 0.4 | 4 | 1+ | 3− | 2+ |
| 0.5 | 7 | 2+ | 4− | 1+ |
| 0.75 | 1 | 4+ | 2− | 1+ |
| 1 | >25 | >25 | 3− | 1+ |
| 1.5 | 14− | 1+ | ||
| 2 | 3− | 2+ | ||
| 2.5 | 13− | 1+ | ||
| 3 | 15− | 1+ | ||
| 3.5 | >25 | 2+ | ||
| 4 | >25 | |||
As can be seen, as the noise level rises, the identification of the pattern set fails, reflecting the necessity of identifying all patterns simultaneously in Equation 3. The statistics gathered on patterns alone still show some recovery of information.